Modelpublic

neko-legends/DeepSeek-V4-Flash-0731-Abliterated-NVFP4

sourceHugging Facemitupdated 1mo agoView on Hugging Face

5likes4downloads

Model Card

DeepSeek V4 Flash 0731 Abliterated NVFP4 — archived conversion

This repo is now an informational page. The checkpoint weights that used to live here have been removed (2026-08-16): the conversion is superseded and there is nothing to download.

In plain English: this was a repackaging of DeepSeek V4 Flash (0731, abliterated) with the routed-expert weights in NVIDIA NVFP4 format. Since then, faster and better-supported paths won out — so instead of these weights, use:

—NVIDIA DGX Spark (GB10): `drowzeys/keys-DeepSeekV4-Flash-GA-0731-Dspark-Abliterated-32-32` — the DSpark-native MXFP4-path checkpoint. Fastest on Sparks in our tests.
—The full 4× Spark TP=4 serving recipe, benches, and fabric runbook: [github.com/neko-legends/spark-bench](https://github.com/neko-legends/spark-bench) — reproduce everything below.
—Server-class Blackwell / RTX PRO stacks: the NVFP4 conversion notes below remain for reference, but the weights are gone; run your own cast with NVIDIA TensorRT Model Optimizer.

What four DGX Sparks do with the 32-32 checkpoint today

Single-stream decode (client wall, 2048-token completions, formal protocol):

C1 decode: TP2 baseline, broken boot, old record, and now

The full ledger — decode, prefill, and concurrency:

The ledger: TP2 vs broken vs record vs now

metric	TP2 baseline (07-31)	TP4 broken, no-spec (08-15)	TP4 record (08-14)	TP4 now (08-16)
C1 decode (tok/s)	67.7	33.5	103.4	136.25 median · 145.5 peak
Prefill cold (tok/s)	1576	~950	~940	2102 @32k · 2156 @8k
C4 aggregate (tok/s)	93.15	92.43	—	182.2

Decode at prompt depth — speculative-decoding acceptance, not depth, is the variable:

decode at depth

Concurrency-4 aggregate:

c4 aggregate

Everything needed to reproduce — image, flags, fabric wiring, boot gates, bench scripts, dated results: [github.com/neko-legends/spark-bench](https://github.com/neko-legends/spark-bench).

About the removed conversion (reference)

—Source: apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8 — routed experts were packed MXFP4 E2M1 with per-32 UE8M0 scales; we used NVIDIA ModelOpt's lossless MXFP4→NVFP4 cast (no dequantize/requantize) with activation scales calibrated on 2× GB10 DGX Spark (128 public prompts, batch 4, seq 512).
—Mixed precision by design: routed experts NVFP4 E2M1; attention/dense FP8 E4M3; shared-expert and MTP tensors retained.
—Export audit at release: 8,657,043,456 / 8,657,043,456 blocks cast, 33,024 tensors across 43 layers, 48/48 shards independently audited, zero structural errors.
—Abliteration retention was verified at release against the upstream ablation manifest (all 36 named tensors bit-identical) plus a small behavioral suite (12/12 substantive, zero refusals).
—The runtime bridge for the SM121 B12X path remains in `runtime/`, the reference inference code in `inference/`, and the DSV4 tokenizer encoding in `encoding/`.

Attribution

—Base model: DeepSeek-AI DeepSeek V4 Flash 0731
—Abliterated source: apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8
—Quantization/export tooling: NVIDIA TensorRT Model Optimizer
—Conversion + serving recipe: neko-legends (spark-bench)

Review the upstream model card and license before deploying any descendant checkpoint.