mlboydaisuke/Gemma-4-31B-CoreAI

Core AI is Apple's on-device ML runtime in iOS 27 / macOS 27 and the successor to Core ML: PyTorch models are exported with Apple's coreai-torch (LLMs: coreai.llm.export) into .aimodel bundles that run on the GPU or the Neural Engine, e.g. Qwen3-8B 4-bit decodes at 94 tok/s on an M4 Max GPU, MLX 90 under the same protocol (apple-silicon-llm-bench, macOS 27 beta 26A5353q, 2026-06-11).

<!-- gen-cards:devicemark begin (managed by scripts/gen-cards + tools/devicemark_row.py — edit cards.json, not this block) --> This model has no row on DeviceMark, the on-device LLM leaderboard. <!-- gen-cards:devicemark end -->

Gemma 4 31B (dense) — Core AI

Apple Core AI (.aimodel) conversion of Google's Gemma 4 31B dense text decoder, ported directly from the QAT release `google/gemma-4-31B-it-qat-q4_0-unquantized`. Decode-only, runs on the stock pipelined engine on Apple Silicon (Mac-class, ~16 GB).

Frontier dense, unblocked by a custom Metal kernel. Gemma 4 31B's full (global) attention layers have a 32-head × 512 Q tensor that overflows MPSGraph's GPU decode scratch heap — the stock SDPA crashes at the first token (apple/coreai-models#27, the same bug as the 12B). This bundle ships a custom flash-decode SDPA kernel on the full layers (block-GQA over the 31B's 4 global KV heads) that removes the offending op, so the model runs.

<!-- gen-cards:use-it begin id=gemma-4-31b (managed by scripts/gen-cards — edit cards.json / QuickStart.swift, not this block) -->

Use it

⚡ One line — run the kit's task op on this model (import CoreAIOps; no session, no model plumbing, downloads on first use):

let tldr = try await CoreAI.summarize(text, options: .model("gemma-4-31b"))

Every op, one shape — Cookbook.

▶️ Run it (source) — the ChatDemo runner (GUI + CLI, one app for every chat model in the catalog):

git clone https://github.com/john-rocky/coreai-kit
open coreai-kit/Examples/ChatDemo/ChatDemo.xcodeproj
# → Run, then pick "Gemma 4 31B" in the model picker

# agents / headless (macOS):
cd coreai-kit/Examples/ChatDemo
swift run chat-cli --model gemma-4-31b --prompt "What can you do, offline?"

💻 Build with it — complete; the glue is kit API, copy-paste runs:

import CoreAIKit

let chat = try await ChatSession(catalog: "gemma-4-31b")
let reply = try await chat.respond(to: prompt)
// reply: the answer, generated fully on-device

When Apple's FoundationModels built-in model isn't enough, keep your session code and swap the model — one line. CoreAIKit's `KitLanguageModel` plugs this bundle into the same system LanguageModelSession; your Tools, @Generable types and transcripts work unchanged, and capabilities (tool calling, guided generation) auto-detect per model.

The take-home is `Examples/ChatDemo/Sources/QuickStart.swift` — this exact code as one typed function, no UI; the CLI is an argument shell over it, and the GUI drives the same ChatSession across turns for its transcript. Multi-turn? Hold the ChatSession and call respond(to:) per turn — it keeps the conversation history; streamResponse(to:) yields tokens as they decode.

Integration checklist

• —SPM: https://github.com/john-rocky/coreai-kit → product CoreAIKit
• —Info.plist: none needed
• —Entitlements: none needed (macOS)
• —First run downloads the model — 20.1 GB (Mac) — then it loads from the local cache (Application Support; progress via the downloadProgress callback)
• —Measure in Release — Debug is ~3× slower on per-token host work <!-- gen-cards:use-it end -->

Bundle (gpu-pipelined/)

int4 from Google's QAT checkpoint (q40 grid). A frontier 31B at int4 is bandwidth-bound, so decode is in the MLX-parity range — the value is "Core AI runs a frontier dense model the stock engine cannot." Mac-only (exceeds the iPhone memory budget). The `g8` suffix is the higher-occupancy flash-decode kernel (8 SIMD-groups per head split the global layers' KV scan; same numerics).

Architecture

Clean dense gemma4 text decoder — no PLE / AltUp / Laurel / MoE / KV-sharing. 60 layers, hidden 5376, 32 heads, vocab 262144, softcap 30, tied embeddings. 5:1 sliding:full; dual headdim (sliding 256 / full `globalheaddim` 512); full layers use `numglobalkeyvalueheads` 4 with `attentionkeqv (value = raw k_proj). Both attention shapes ride one growing KV pair, so the bundle loads on the stock CoreAIPipelinedEngine` (2 states, no engine patch); the full layers' SDPA runs as a custom Metal flash-decode kernel.

Usage

huggingface-cli download mlboydaisuke/Gemma-4-31B-CoreAI \
    --include "gpu-pipelined/gemma4_31b_qat_decode_int4linsym_msdpa_g8/*" \
    --local-dir ./gemma4-31b-coreai

COREAI_CHUNK_THRESHOLD=1 llm-runner \
    --model ./gemma4-31b-coreai/gpu-pipelined/gemma4_31b_qat_decode_int4linsym_msdpa_g8 \
    --prompt "What is the capital of France?" --max-tokens 64 --chunk-size 1

Conversion

Community zoo: github.com/john-rocky/coreai-model-zoo → `zoo/gemma4-31b.md`.

License

Gemma 4 is released under the Apache License 2.0 — see the base card google/gemma-4-31B-it-qat-q4_0-unquantized and Google's Gemma 4 license page. The conversion (Core AI bundles) adds no additional restrictions.

<!-- funnel:v1 -->

More models in this format: Core AI Model Zoo — 75 models, each with the recipe that produced it.

Want a different model on-device? Open a request — free, open weights only; the export and its measured numbers get published publicly.

<!-- /funnel:v1 -->