Modelpublic

KIEFERSA/Sophea-Titan-1

sourceHugging Faceapache-2.0updated 12d agoView on Hugging Face

2likes51downloads

Model Card

<div style="display:flex;align-items:center;gap:18px;padding:16px 20px;margin-bottom:16px;border-radius:14px;background:linear-gradient(100deg,#e8f5ec 0%,#f7fbf8 62%);border:1px solid #c6e7d1"><img alt="KIEFERSA" style="flex:0 0 auto;height:58px;width:58px;border-radius:12px;object-fit:contain" src="data:image/webp;base64,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"/><div><div style="font-size:1.24rem;font-weight:700;color:#003713;line-height:1.2">Sophea-Titan-1</div><div style="font-size:.86rem;color:#3a5a47;margin-top:2px">Greek fine-tuned multimodal Qwen3.6-27B — non-thinking chat</div></div></div>

Sophea-Titan-1 is a 27B Greek fine-tuned multimodal chat model built on Qwen3.6-27B, with register control and a fixed assistant identity — while retaining English ability and vision.

—Creator: Kiefer SA
—Base model: Qwen3.6-27B (multimodal, 64 layers, hidden 5120)
—Languages: Greek (primary), English (retained)
—Decoding: non-thinking (enable_thinking=false) — see recommended sampling under Usage

Serve non-thinking. Set enable_thinking=false in the chat template. Over an OpenAI-compatible vLLM endpoint, pass extra_body={"chat_template_kwargs": {"enable_thinking": false}}. With thinking left on, Greek output quality degrades sharply.

Intended use

—General-purpose Greek conversational assistant / chat, with formal ↔ informal register control
—Greek-knowledge QA; English retained as a secondary language

Not evaluated for safety-critical or legal/medical decisions.

Fine-tuned from Qwen3.6-27B.

Evaluation

Non-thinking, greedy (temperature 0). Sophea-Titan-1 in bold.

Models compared:

n/a = not applicable (text-only) · n.t. = not identity-tuned.

Headline scorecard

At a glance

Capability profile — General Greek benchmarks, English and vision across open-weight models (Sophea-Titan-1 in bold; frontier MoEs GLM-5.2 / MiniMax-M3 shown for reference, no vision):

[image]

General Greek benchmarks vs. model size — Sophea-Titan-1 has the strongest general-Greek score of the ~27B open models, approaching far larger frontier MoEs at a fraction of the size:

[image]

Per-benchmark detail

Three benchmark families in one table — General Greek benchmarks (9), English retention (5), and Vision — MMStar (1500) — on the same five models. n/a = not applicable (text-only, no vision).

<div style="overflow-x:auto"> <table style="width:100%;border-collapse:collapse;font-size:13px;font-variant-numeric:tabular-nums;font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif"> <thead><tr><th style="padding:10px 8px;text-align:left;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Benchmark</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Base</th><th style="padding:10px 8px;text-align:center;font-weight:700;border-bottom:2px solid #0a8043;color:#00682f;font-size:14px;background:rgba(10,128,67,.10);">Sophea-Titan-1</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Sophea-K1</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3-30B-A3B</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3.6-35B-A3B</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Krikri-8B</th></tr></thead><tbody><tr><td colspan="7" style="padding:10px 10px 8px;font-weight:700;color:#00682f;background:rgba(10,128,67,.12);border-bottom:2px solid #0a8043;font-size:13px">General Greek benchmarks (9)</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">greekmmlu</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.849</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.854</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.855</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.753</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.838</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.675</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">mmlugreek</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.798</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.797</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.782</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.628</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.770</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.519</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">hellaswag</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.611</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.680</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.622</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.410</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.574</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.572</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">medicalmcqa</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.312</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.387</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.368</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.232</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.312</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.287</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">winogrande</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.590</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.626</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.599</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.559</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.564</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.609</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">arcchallenge</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.944</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.950</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.945</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.867</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.916</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.690</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">arceasy</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.973</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.973</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.971</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.931</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.969</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.832</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">belebele</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.941</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.950</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.938</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.887</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.926</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.771</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">truthfulqa</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.423</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.415</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.437</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.446</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.416</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.425</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">MACRO (9)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.7157</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.7369</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.7240</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.6346</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.6983</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.5977</td></tr><tr><td colspan="7" style="padding:10px 10px 8px;font-weight:700;color:#00682f;background:rgba(10,128,67,.12);border-bottom:2px solid #0a8043;font-size:13px">English retention (5)</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">arcchallenge</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.972</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.979</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.977</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.932</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.959</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.765</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">arceasy</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.990</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.992</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.994</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.981</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.990</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.892</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">hellaswag</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.764</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.805</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.780</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.500</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.734</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.751</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">mmlu</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.858</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.858</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.858</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.745</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.840</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.608</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">winogrande</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.732</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.758</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.742</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.658</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.658</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.695</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">MACRO (5)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8631</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.8783</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8702</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.7631</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8362</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.7423</td></tr><tr><td colspan="7" style="padding:10px 10px 8px;font-weight:700;color:#00682f;background:rgba(10,128,67,.12);border-bottom:2px solid #0a8043;font-size:13px">Vision — MMStar (1500)</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">coarse perception</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.744</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.728</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.728</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.704</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">fine-grained perception</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.636</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.608</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.628</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.588</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">instance reasoning</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.756</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.792</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.792</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.740</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">logical reasoning</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.668</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.764</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.636</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.572</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">math</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.464</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.708</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.500</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.372</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">science & technology</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.652</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.684</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.688</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.564</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">OVERALL</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.653</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.714</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.662</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.590</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">n/a</td></tr></tbody> </table></div>

greekmmlu — per-subject (31)

<div style="overflow-x:auto"> <table style="width:100%;border-collapse:collapse;font-size:13px;font-variant-numeric:tabular-nums;font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif"> <thead><tr><th style="padding:10px 8px;text-align:left;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">subject</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Base</th><th style="padding:10px 8px;text-align:center;font-weight:700;border-bottom:2px solid #0a8043;color:#00682f;font-size:14px;background:rgba(10,128,67,.10);">Sophea-Titan-1</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Sophea-K1</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3-30B-A3B</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3.6-35B-A3B</th><th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Krikri-8B</th></tr></thead><tbody><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Accounting</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.864</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.864</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.859</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.761</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.848</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.663</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Agriculture</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.853</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.870</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.853</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.721</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.819</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.707</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Art</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.748</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.756</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.770</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.621</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.757</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.625</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Biology</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.864</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.868</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.859</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.794</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.856</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.672</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Chemistry</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.827</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.753</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.778</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.654</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.790</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.519</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Civil Engineering</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.793</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.827</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.819</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.678</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.785</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.637</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Clinical Knowledge</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.810</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.820</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.795</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.689</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.790</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.686</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Computer Networks & Security</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.730</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.762</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.714</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.540</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.667</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.476</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Computer Science</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.883</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.880</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.866</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.827</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.869</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.768</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Driving Rules</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.837</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.820</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.829</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.756</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.811</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.631</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Economics</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.911</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.926</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.926</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.840</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.894</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.670</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Education</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.857</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.867</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.895</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.762</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.901</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.687</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Electrical Engineering</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.830</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.822</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.844</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.681</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.817</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.538</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">General Knowledge</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.769</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.792</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.823</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.701</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.781</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.630</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Geography</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.943</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.964</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.955</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.870</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.950</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.872</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Government and Politics</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.944</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.941</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.961</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.901</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.944</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.899</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Greek History</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.871</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.877</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.878</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.711</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.884</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.816</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Greek Literature</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.714</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.714</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.643</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.429</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.500</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.500</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Greek Mythology</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.870</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.857</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.845</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.744</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.849</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.697</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Greek Traditions</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.880</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.891</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.880</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.755</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.867</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.710</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Law</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.718</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.703</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.697</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.580</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.677</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.504</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Management</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.824</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.841</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.842</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.719</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.814</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.694</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Maritime Safety & Rescue</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.703</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.669</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.662</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.608</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.682</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.507</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Mathematics</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.897</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.924</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.904</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.844</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.857</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.516</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Medicine</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.893</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.895</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.886</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.757</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.882</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.663</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Modern Greek Language</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.916</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.918</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.928</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.836</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.911</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.780</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Physics</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.838</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.851</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.851</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.769</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.838</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.684</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Prehistory</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">1.000</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">1.000</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.968</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.984</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.984</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.905</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">World History</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.950</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.950</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.900</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.850</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.950</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.800</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">World Religions</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.768</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.755</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.787</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.652</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.800</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.684</td></tr><tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">OVERALL</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.849</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.854</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.855</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.753</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.838</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.675</td></tr></tbody> </table></div>

Frontier & other API models (reference)

Evaluated via API (thinking disabled where supported): GPT-5.5, Claude Opus 4.8, Gemini 3.5, GLM-5.2, MiniMax-M3. API-model Greek/English benchmarks use letter-answer scoring (local models use log-likelihood). Kimi-K3 added 2026-07-29: kimi-k3 via Moonshot API (temperature=1 forced, thinking cannot be disabled). Inkling-Small added 2026-07-31: thinkingmachines/Inkling-Small-NVFP4 (276B/12B-active open-weight MoE, Apache-2.0), self-hosted via vLLM; thinking on, temperature=1 (same protocol caveat as noted for reasoning-locked APIs). Qwen3.8-Max added 2026-08-04: qwen3.8-max via Qwen Cloud API (thinking disabled, temperature=0 — matches the card protocol). DeepSeek-V4.1-Flash added 2026-09-14: deepseek-ai/DeepSeek-V4.1-Flash (552B/16B-active open-weight MoE + 196B Engram tables, MIT), self-hosted via vLLM on 3× B200 (FP4 experts / FP8 dense as shipped); thinking disabled, temperature=0 — matches the card protocol. 4 of 7,000 answers were unparsed (the model reasoned in-content instead of answering with a letter) and are scored as wrong. Vision — MMStar (1500, same letter-answer protocol, thinking disabled): <strong>0.721</strong> overall (coarse perception 0.756 · fine-grained perception 0.620 · instance reasoning 0.768 · logical reasoning 0.764 · math 0.800 · science & technology 0.620; 11 unparsed scored as wrong) vs Sophea-Titan-1 0.714.

<th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3.8-Max</th> <th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">DeepSeek-V4.1-Flash</th> </tr></thead><tbody> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">General Greek benchmarks macro (9)</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.737</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.909</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.922</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.872</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.810</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.821</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.916</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.881</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.845</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">English macro (5)</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.878</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.925</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.940</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.911</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.892</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.890</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.947</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.922</td>

General Greek benchmarks (9)

<th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3.8-Max</th> <th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">DeepSeek-V4.1-Flash</th> </tr></thead><tbody> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">greekmmlu</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.854</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.905</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.900</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.898</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.805</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.837</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.902</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.872</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.858</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">mmlu_greek</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.797</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.888</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.884</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.874</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.769</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.773</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.932</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.898</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.846</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">hellaswag</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.680</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.891</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.909</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.816</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.693</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.741</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.790</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.666</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.638</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">medical_mcqa</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.387</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.928</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.917</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.912</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.787</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.833</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.944</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.926</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.860</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">winogrande</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.626</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.781</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.813</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.641</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.705</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.662</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.880</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.850</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.734</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">arc_challenge</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.950</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.968</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.966</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.952</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.904</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.924</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.972</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.960</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.962</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">belebele</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.950</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.953</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.947</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.934</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.913</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.911</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.956</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.954</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.944</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">truthfulqa</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.415</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.885</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.977</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.845</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.766</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.749</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.882</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.832</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.824</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">MACRO</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.7369</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9094</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9216</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8717</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8102</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8212</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9158</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8811</td>

English — 5 benchmarks

<th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3.8-Max</th> <th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">DeepSeek-V4.1-Flash</th> </tr></thead><tbody> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">arc_challenge</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.979</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.976</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.974</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.958</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.968</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.967</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.978</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.968</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.984</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">hellaswag</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.805</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.918</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.944</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.913</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.858</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.854</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.876</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.806</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.894</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">winogrande</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.758</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.830</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.877</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.811</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.801</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.793</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.946</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.922</td>

<td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.760</td> </tr> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">MACRO</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.8783</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9249</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9399</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9113</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8922</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.8903</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9472</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.9224</td>

greekmmlu

<th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Qwen3.8-Max</th> <th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">DeepSeek-V4.1-Flash</th> </tr></thead><tbody> <tr> <td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">greekmmlu</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;color:#00682f;background:rgba(10,128,67,.10);">0.854</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.905</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.900</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.898</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.805</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.837</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.902</td> <td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">0.872</td>

Quantized variants & other formats

Smaller-footprint and alternate-runtime builds of this model:

—GPU / vLLM (text-only — vision tower not included): **Sophea-Titan-1-FP8** — lossless, ≈29 GB · **Sophea-Titan-1-NVFP4** — 4-bit, ≈19 GB
—llama.cpp — **Sophea-Titan-1-GGUF**
—Apple Silicon — **Sophea-Titan-1-mlx**

The trade-off plot below covers the vLLM builds scored on the Greek suite; GGUF / MLX are format conversions.

[image]

<p style="text-align:center;font-size:.82rem;color:#5a7a67;font-style:italic;margin:2px 0 16px">VRAM vs. Greek benchmark score across the Sophea-Titan-1 quantized family (bf16 → FP8 → NVFP4).</p>

Usage

Serve with vLLM (OpenAI-compatible; 27B bf16 ≈ 54 GB — add --tensor-parallel-size N for multi-GPU):

bash

vllm serve KIEFERSA/Sophea-Titan-1 --served-model-name sophea-titan-1 --trust-remote-code \
  --enable-auto-tool-choice --tool-call-parser qwen3_coder \
  --reasoning-parser qwen3

For text-only serving (skips the vision tower — much faster), add --language-model-only.

Serve with vision enabled — raise the per-request image budget explicitly:

bash

vllm serve KIEFERSA/Sophea-Titan-1 --served-model-name sophea-titan-1 --trust-remote-code \
  --limit-mm-per-prompt '{"image": 4}' \
  --enable-auto-tool-choice --tool-call-parser qwen3_coder \
  --reasoning-parser qwen3

python

resp = client.chat.completions.create(
    model="sophea-titan-1",
    messages=[{"role": "user", "content": [
        {"type": "image_url", "image_url": {"url": "https://example.com/chart.png"}},
        {"type": "text", "text": "Περίγραψε την εικόνα στα ελληνικά."},
    ]}],
    temperature=0,
    extra_body={"chat_template_kwargs": {"enable_thinking": False}},   # required: non-thinking
)
print(resp.choices[0].message.content)

Recommended sampling (instruct / non-thinking): temperature=0.7, top_p=0.80, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0.

Client (OpenAI SDK) — remember non-thinking:

python

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
resp = client.chat.completions.create(
    model="sophea-titan-1",
    messages=[{"role": "user", "content": "Ποια είναι η πρωτεύουσα της Ελλάδας;"}],
    temperature=0,
    extra_body={"chat_template_kwargs": {"enable_thinking": False}},   # required: non-thinking
)
print(resp.choices[0].message.content)

Transformers (multimodal — load with the image-text-to-text head to keep vision):

python

import torch
from transformers import AutoModelForImageTextToText, AutoProcessor

proc = AutoProcessor.from_pretrained("KIEFERSA/Sophea-Titan-1", trust_remote_code=True)
model = AutoModelForImageTextToText.from_pretrained(
    "KIEFERSA/Sophea-Titan-1", torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)

messages = [{"role": "user", "content": [{"type": "text",
             "text": "Ποια είναι η πρωτεύουσα της Ελλάδας;"}]}]
text = proc.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)
inputs = proc(text=[text], return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=256, do_sample=False)
print(proc.batch_decode(out[:, inputs.input_ids.shape[1]:], skip_special_tokens=True)[0])

Speculative decoding (MTP)

This model ships the multi-token-prediction head — 15 mtp.* tensors (~0.85 GB, bf16) in model-mtp.safetensors. This is the single-layer draft stack that config.json has always declared through text_config.mtp_num_hidden_layers: 1.

Earlier revisions did not contain it. The LoRA merge loaded the base through AutoModelForCausalLM, and transformers declares _keys_to_ignore_on_load_unexpected = [r"^mtp.*"] for this architecture, so the head was discarded at load and never written back — the config advertised a module the weights did not contain. The head has been restored tensor-for-tensor from the base, leaving every other weight untouched; the addition is purely additive, so existing serve commands keep working unchanged. To pin the previous bytes, use revision="726f410a802dcf7a720dca898da9c7b91943911e".

Enable it with vLLM (≥ 0.23.0):

bash

vllm serve KIEFERSA/Sophea-Titan-1 --served-model-name sophea-titan-1 --trust-remote-code \
  --speculative-config '{"method":"qwen3_5_mtp","num_speculative_tokens":1}'

Provenance. These are the base Qwen3.6-27B MTP weights. The Greek SFT and DPO LoRA never targeted mtp.*, so no fine-tuned draft head exists. This cannot affect output quality: speculative decoding verifies every drafted token against the main model, so a stale drafter changes throughput only, never the output distribution.

The -FP8 and -NVFP4 variants ship the same head and support the same flag; on those builds the MTP Linears are held in bf16 and listed in quantization_config.ignore.

License

Inherits the Qwen3.6-27B base-model license. Verify base-model terms before use.