bytkim / text-generation updated 3 months ago

Qwen3.6-27B-MTP-pi-tune-GGUF

No-Thinking · PI Tune · MTP · GGUF 🪐 Qwen3.6-27B-MTP-pi-tune A 4-bit QLoRA SFT Multi-Token Prediction tune of Qwen3.6-27B fine-tuned for no-thinking agentic coding through a PI-style harness. Packaged as llama.cpp-compatible GGUF for local agent loops. 🧠 27B Dense Foundation 🚫 No-Thinking Tuned ⚡ MTP Speculative Decoding 🛠️ Coding · DevOps · Agents 📦 llama.cpp GGUF 🖼️ Multi-Modal Compatible 🪟 256k Native · 1M M...

Params
Context
Downloads 30d
319 K
Likes
127
Commercial use: allowed apache-2.0 Not gated GGUF View on Hugging Face ↗

Download history

daily snapshots · 31 days
▲ 131 K in the last 30 days (69.3%)
331 K188 K
Aug 22Sep 1Sep 11Sep 21

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
Qwen3.6-27B-MTP-pi-tune-Q2_K.gguf Q2_K 10.9 GB 12.5 GB ✅ Runs comfortably
Qwen3.6-27B-MTP-pi-tune-Q3_K_S.gguf Q3_K_S 12.3 GB 14.0 GB ✅ Runs comfortably
Qwen3.6-27B-MTP-pi-tune-Q3_K_M.gguf Q3_K_M 13.5 GB 15.4 GB ✅ Runs comfortably
Qwen3.6-27B-MTP-pi-tune-Q3_K_L.gguf Q3_K_L 14.6 GB 16.6 GB ✅ Runs comfortably
Qwen3.6-27B-MTP-pi-tune-Q4_K_S.gguf Q4_K_S 15.9 GB 17.9 GB ✅ Runs comfortably
Qwen3.6-27B-MTP-pi-tune-Q4_K_M.gguf Q4_K_M 16.8 GB 19.0 GB ✅ Runs comfortably
Qwen3.6-27B-MTP-pi-tune-Q5_K_S.gguf Q5_K_S 19.0 GB 21.4 GB ⚠️ Tight — reduce context
Qwen3.6-27B-MTP-pi-tune-Q5_K_M.gguf Q5_K_M 19.6 GB 22.0 GB ⚠️ Tight — reduce context
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub
~ · ollama · Q4_K_M
$ ollama run qwen3-6-27b-mtp-pi-tune-gguf

# pin the quantization explicitly
$ ollama run qwen3-6-27b-mtp-pi-tune-gguf-q4_k_m
est. VRAM 19.0 GBon RTX 4090 · 24 GBJSON API →

Specifications

Licence
apache-2.0
First seen on the Hub
2026-06-02
Base model
Qwen3.6-27B
Training datasets
undisclosed
Terminal-Bench 2.0 (reported)
28.09
Added to our catalog
2026-08-22

Family

Base model and the most-downloaded derivatives in the catalog.

Compare with any text-generation model