FINAL-Bench/POCKET-35B-GGUF

### 🆕 [POCKET-Qwen3.8-Flash-Next](https://huggingface.co/FINAL-Bench/POCKET-Qwen3.8-Flash-Next-GGUF) — a 180B model running on a laptop with 8 GB VRAM + 32 GB RAM · 4.17 tok/s measured.  ![VRAM]() ![RAM]() ![Speed]() <!-- POCKET-FLASHNEXT-BADGE -->

### 🆕 [POCKET-Zimage-CPU](https://huggingface.co/FINAL-Bench/POCKET-Zimage-CPU) — photoreal images in 46 s on a CPU only. No GPU, no CUDA, no Python.   ![RAM]() <!-- POCKET-Zimage-CPU-BADGE -->

### 📚 Collections ▶ [POCKET Models](https://huggingface.co/collections/FINAL-Bench/pocket-models-6a618ee5d23eafb7e185a5c6) — this family (on-device, no GPU) Darwin Family · Aether Foundation · VKAE Accelerated

[image]

POCKET-35B-GGUF

A 35B model that runs on your PC with no GPU — and on your phone. Just stock llama.cpp. No fork, no CUDA, no cloud.

🚀 Try it live, no install →   — both answering on a CPU-only box (no GPU). POCKET-26B is Gemma4-based.

  ![No GPU]() ![Base]()

Pick your build →        

The POCKET lineup — pick by your device

*Wikipedia-Korean perplexity, lower is better. Q4_K_M = 5.79 baseline. English builds are tuned on English; see each repo. †Separate 80-chunk run (40,960 tokens) on a different model — compare within a model, not across rows.

🍎 Why MLX for Korean but GGUF for English on iPhone? Apple-native MLX only does uniform quantization. Korean survives it; English needs our proprietary quantization, which only GGUF supports — so the English iPhone build ships as a GGUF you run with PocketPal. Honest, not lazy.

🆕 POCKET-26B — a Gemma4-26B-A4B-based sibling that loads in any app today (Ollama · LM Studio · PocketPal · MLX), no bleeding-edge runtime needed: [GGUF](https://huggingface.co/FINAL-Bench/POCKET-26B-GGUF) (Q2_K 11 GB · Q4_K_M 17 GB · GPQA-Diamond 67%). Universal compatibility for 12 GB phones, PC, and browser.

[image]

Benchmarks — what is measured, what is not

We measure Bonsai on the same machine with the same stock `llama.cpp`, and we tell you where we lose.

[measured] Generation speed — POCKET wins on both CPU and GPU:

[measured on a MacBook M3 Pro, 18 GB] — and on a laptop, POCKET wins every axis, including prompt processing:

On a laptop GPU the arithmetic headroom that let Bonsai win prefill on an H100 is gone, so MoE sparsity wins across the board. POCKET-35B-Q2_K runs on the M3 Pro's CPU at 19.5 tok/s — on an 18 GB Mac, run Q2_K on CPU (-ngl 0); its 13 GB exceeds the recommended Metal budget.

[measured — GPQA Diamond, 198q, greedy] reasoning quality vs quantization:

[pending — community reports welcome] on-device iPhone and Strix Halo throughput. We publish only what we ran ourselves; help us fill the rest.

The same-size rival Ternary-Bonsai-27B-Q2_0 (7.2 GB) fails to load in upstream llama.cpp — it needs the PrismML fork. POCKET runs on the tools you already have.

Files in this repo

Quickstart — no fork needed

# any recent llama.cpp — brew / winget / apt, or LM Studio / Ollama
llama-cli -m POCKET-35B-Q2_K.gguf -p "안녕하세요" -ngl 0 -t 8
# reproduce our CPU numbers:
llama-bench -m POCKET-35B-IQ1_M.gguf -p 128 -n 64 -ngl 0 -t 16

Use physical-core count for -t (max ~32). Do not pass all threads — it can slow down sharply.

Lineage — where POCKET comes from

POCKET is quantized from [Darwin-36B-Opus](https://huggingface.co/FINAL-Bench/Darwin-36B-Opus), VIDRAFT's flagship — a model bred and evolved over several generations on the Darwin platform (crossbreeding, healing, expert surgery). Darwin-36B-Opus itself traces back to a Qwen3.5-family MoE architecture.

The CPU/GPU speed comes from the sparse-MoE architecture plus ordinary quantization — reproducible with the same base and the same tools. What we add is the Darwin-evolved weights, the honest measurement, the Korean tuning, and the pruning that makes the 5 GB phone builds.

Limitations

• —The iPhone/Mac speed is not yet measured by us — community reports welcome.
• —Extreme quants (IQ1_M) hurt Korean ~2.8× more than English; use Q2_K or larger for quality.
• —English phone builds trade quality for size; the PC build (PC-mix) is much closer to full quality.

License

Apache-2.0.

POCKET is a VIDRAFT model family. 35B, in your pocket. No GPU.

Learn more

• —On-device LLMs without a GPU — and how POCKET measures up: Can you run a large LLM without a GPU?
• —What model quantization is, and why a 4-bit model stays smart: What is model quantization?

<!-- POCKET-FAMILY -->

🧩 The POCKET Family — On-device AI by VIDRAFT

Big models, small hardware. No GPU, no cloud.

Models

• —📦 POCKET-35B-GGUF — flagship, PC / server, no GPU
• —📦 POCKET-26B-GGUF — compact 26B
• —🇰🇷 POCKET-KR-GGUF — Korean, Android
• —🍎 POCKET-KR-MLX — Korean, iPhone / Mac
• —🌍 POCKET-EN-GGUF — English, phone / PC
• —💻 POCKET-Qwen3.8-Flash-Next-GGUF — 180B on a laptop (8 GB VRAM + 32 GB RAM)
• —🖼️ POCKET-Image-Zimage — character-perfect text in any image
• —🖥️ POCKET-Zimage-CPU — photoreal images on a CPU only

Demos & tools (Spaces)

• —🎨 POCKET-Image Studio — text-in-image, generate in-page
• —🖥️ POCKET-35B-CPU — 35B answering on a CPU
• —🖥️ POCKET-26B-CPU — 26B on a CPU
• —🖼️ POCKET-Zimage-CPU — image generation on a CPU

📚 Full POCKET collection

<!-- /POCKET-FAMILY -->