Article
The Fast Gemma Challenge: our verified-SOTA recipe, in full
FINAL-Bench
• • 21
Warecube Korea
Warecube
AI & ML interests
None yet
Recent Activity
upvoted an article 3 days ago
The Fast Gemma Challenge: our verified-SOTA recipe, in full reacted to SeaWolf-AI's post with 🚀 3 days ago
We wrote up our run in The Fast Gemma Challenge — as vidraft-darwin — and wanted to share the recipe. 🙏
https://huggingface.co/spaces/gemma-challenge/gemma-dashboard
Verified result: 510.58 TPS at PPL 2.3930 on a single A10G (fw188-ctk49-n64-patchbridge, re-run & VERIFIED). Honest note: on raw TPS there are faster runs (535+), but those went over the PPL bar and didn't verify — what we're proud of is the fastest result that keeps quality.
The recipe is already open, so we explained each piece: sliding-window W188, CTK49 kernel tuning, noprecache (honest, verifiable measurement), and an N64 synthetic warmup bridge that shrinks the public↔private gap (~15 TPS), plus INT4 + MTP K=7 + CUDA-graph capture. One rule: only stack quality-neutral speedups.
Huge thanks to @firfir-cast, @gemma-slayer, @chiku-inu, @kenyan-duma, @dixie-flatline and everyone who shared their experiments. Full write-up
👇
https://huggingface.co/blog/FINAL-Bench/fast-gemma
liked a model 11 days ago
FINAL-Bench/POCKET-Image-ZimageOrganizations
None yet
upvoted an article 3 days ago
reacted to SeaWolf-AI's post with 🚀 3 days ago
Post
3347
We wrote up our run in The Fast Gemma Challenge — as vidraft-darwin — and wanted to share the recipe. 🙏
gemma-challenge/gemma-dashboard
Verified result: 510.58 TPS at PPL 2.3930 on a single A10G (fw188-ctk49-n64-patchbridge, re-run & VERIFIED). Honest note: on raw TPS there are faster runs (535+), but those went over the PPL bar and didn't verify — what we're proud of is the fastest result that keeps quality.
The recipe is already open, so we explained each piece: sliding-window W188, CTK49 kernel tuning, noprecache (honest, verifiable measurement), and an N64 synthetic warmup bridge that shrinks the public↔private gap (~15 TPS), plus INT4 + MTP K=7 + CUDA-graph capture. One rule: only stack quality-neutral speedups.
Huge thanks to @firfir-cast , @gemma-slayer , @chiku-inu , @kenyan-duma , @dixie-flatline and everyone who shared their experiments. Full write-up
👇
https://huggingface.co/blog/FINAL-Bench/fast-gemma
gemma-challenge/gemma-dashboard
Verified result: 510.58 TPS at PPL 2.3930 on a single A10G (fw188-ctk49-n64-patchbridge, re-run & VERIFIED). Honest note: on raw TPS there are faster runs (535+), but those went over the PPL bar and didn't verify — what we're proud of is the fastest result that keeps quality.
The recipe is already open, so we explained each piece: sliding-window W188, CTK49 kernel tuning, noprecache (honest, verifiable measurement), and an N64 synthetic warmup bridge that shrinks the public↔private gap (~15 TPS), plus INT4 + MTP K=7 + CUDA-graph capture. One rule: only stack quality-neutral speedups.
Huge thanks to @firfir-cast , @gemma-slayer , @chiku-inu , @kenyan-duma , @dixie-flatline and everyone who shared their experiments. Full write-up
👇
https://huggingface.co/blog/FINAL-Bench/fast-gemma
- 8 replies
liked a
model 11 days ago
liked a
Space 11 days ago
reacted to SeaWolf-AI's post with 🧠🔥 12 days ago
Post
3124
POCKET now speaks Gemma 4 — a 26B model that loads in every app, and runs on your PC with no GPU
We're adding a Gemma-4 sibling to POCKET: POCKET-26B, built from Google's Gemma-4-26B-A4B (Apache-2.0). Our flagship POCKET-35B is a Qwen-family MoE and needs a recent llama.cpp; POCKET-26B trades a little size for the thing people kept asking for — it just loads, everywhere, today: Ollama, LM Studio, PocketPal, MLX, any stock llama.cpp. No fork, no bleeding-edge runtime, no CUDA, no cloud.
It's a sparse Mixture-of-Experts (25.2B total, ~4B active per token), so the work per token stays small — a real 26B that generates on a CPU with no graphics card.
Two things make it stand out:
1) Universal compatibility. Gemma 4 is a standard, widely-supported architecture, so POCKET-26B runs on the tools you already have — no waiting for your app to add a new model type.
2) Quality that survives compression. Measured GPQA-Diamond (198 q, greedy):
• Full base: 67.7%
• POCKET-26B Q4_K_M (17 GB): 67.7% — lossless
• POCKET-26B Q2_K (11 GB): 67.2% — near-lossless, at 11 GB
Live, on a CPU-only box (our demo Space — POCKET-26B vs Bonsai-27B, same machine, same stock llama.cpp): POCKET-26B ≈ 19 tok/s vs Bonsai ≈ 6 tok/s → about 3× faster generation, no GPU. (Honest notes: shared CPU box, sequential race; a dedicated machine is faster.)
Where it fits in the family:
• POCKET-35B (Qwen MoE) — bigger, top-tier, needs a recent llama.cpp.
• POCKET-26B (Gemma 4) — loads in any app, quality-robust when compressed. The demo runs the Q4_K_M build; Q2_K (11 GB) is the smallest footprint. For a true ≤8 GB phone, the 5 GB POCKET-KR (Qwen) is still the pick.
Try it and grab it:
🖥️ Live demo (Gemma4-based, answering on a CPU, no GPU): FINAL-Bench/POCKET-26B-CPU
📦 POCKET-26B-GGUF (Q4_K_M 17 GB · Q2_K 11 GB): FINAL-Bench/POCKET-26B-GGUF
📚 POCKET collection: https://huggingface.co/collections/FINAL-Bench/pocket-models
We're adding a Gemma-4 sibling to POCKET: POCKET-26B, built from Google's Gemma-4-26B-A4B (Apache-2.0). Our flagship POCKET-35B is a Qwen-family MoE and needs a recent llama.cpp; POCKET-26B trades a little size for the thing people kept asking for — it just loads, everywhere, today: Ollama, LM Studio, PocketPal, MLX, any stock llama.cpp. No fork, no bleeding-edge runtime, no CUDA, no cloud.
It's a sparse Mixture-of-Experts (25.2B total, ~4B active per token), so the work per token stays small — a real 26B that generates on a CPU with no graphics card.
Two things make it stand out:
1) Universal compatibility. Gemma 4 is a standard, widely-supported architecture, so POCKET-26B runs on the tools you already have — no waiting for your app to add a new model type.
2) Quality that survives compression. Measured GPQA-Diamond (198 q, greedy):
• Full base: 67.7%
• POCKET-26B Q4_K_M (17 GB): 67.7% — lossless
• POCKET-26B Q2_K (11 GB): 67.2% — near-lossless, at 11 GB
Live, on a CPU-only box (our demo Space — POCKET-26B vs Bonsai-27B, same machine, same stock llama.cpp): POCKET-26B ≈ 19 tok/s vs Bonsai ≈ 6 tok/s → about 3× faster generation, no GPU. (Honest notes: shared CPU box, sequential race; a dedicated machine is faster.)
Where it fits in the family:
• POCKET-35B (Qwen MoE) — bigger, top-tier, needs a recent llama.cpp.
• POCKET-26B (Gemma 4) — loads in any app, quality-robust when compressed. The demo runs the Q4_K_M build; Q2_K (11 GB) is the smallest footprint. For a true ≤8 GB phone, the 5 GB POCKET-KR (Qwen) is still the pick.
Try it and grab it:
🖥️ Live demo (Gemma4-based, answering on a CPU, no GPU): FINAL-Bench/POCKET-26B-CPU
📦 POCKET-26B-GGUF (Q4_K_M 17 GB · Q2_K 11 GB): FINAL-Bench/POCKET-26B-GGUF
📚 POCKET collection: https://huggingface.co/collections/FINAL-Bench/pocket-models
- 3 replies
liked a
Space 12 days ago
liked 10
models 12 days ago
FINAL-Bench/POCKET-26B-GGUF
Text Generation • 25B • Updated • 27.3k • 39
mradermacher/Darwin-V9-Chimera-4B-i1-GGUF
4B • Updated • 1.09k • 7
mradermacher/Darwin-4B-Chimera-i1-GGUF
4B • Updated • 800 • 7
mradermacher/Darwin-V9-Chimera-4B-SFT-i1-GGUF
4B • Updated • 1.36k • 7
bartowski/FINAL-Bench_Darwin-4B-Chimera-GGUF
Text Generation • 4B • Updated • 2.15k • 7
Pq234/Darwin-28B-Opus
Text Generation • 28B • Updated • 388 • 6
mradermacher/Darwin-4B-Chimera-GGUF
4B • Updated • 576 • 7
usermma/Darwin-28B-Coder-mlx-fp16
Text Generation • 27B • Updated • 167 • 7
el4/Darwin-36B-Opus-OPAL-GGUF
Text Generation • 35B • Updated • 13.9k • 15
TheCluster/Darwin-31B-Opus-MLX-8bit
Image-Text-to-Text • 33B • Updated • 175 • 12
upvoted a paper 14 days ago
upvoted an article 14 days ago
Article
POCKET: a 35-billion-parameter model that runs on your iPhone — and on your PC with no GPU
FINAL-Bench
• • 12 liked a
Space 14 days ago