HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen3.6-27B

Qwen/Qwen3.6-27B

family qwen · 26 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen3.6-27B
best scores
Terminal-Bench: 2.0MMLU-Pro: 86.25GPQA: 86.34AIME 2025: 93.12QuALITY: 100.0

Best scores per benchmark

benchmarkcategorybest score
QuALITYknowledge100.0
AIME 2025math93.12
GPQAreasoning86.34
MMLU-Proknowledge86.25
Terminal-Benchagentic-coding2.0

Variants (26)

variantkindquantbitsvRAMdlorgyear
Qwen3.6-27B-FP8fp8fp88.07,072,045Qwen2026
Qwen3.6-27Boriginalbf1616.06,101,101Qwen2026
Qwen3.6-27B-MTP-GGUFggufgguf16.03,006,469unsloth2026
Qwen3.6-27B-NVFP4nvfp4nvfp44.02,668,382unsloth2026
Qwen3.6-27B-NVFP4nvfp4nvfp44.01,915,154nvidia2026
Qwen3.6-27B-MLX-8bitmlxmlx1,025,409lmstudio-community2026
Qwen3.6-27B-MLX-4bitmlxmlx973,375lmstudio-community2026
Qwen3.6-27B-MLX-6bitmlxmlx942,405lmstudio-community2026
Qwen3.6-27B-MLX-5bitmlxmlx923,303lmstudio-community2026
Qwen3.6-27B-GGUFggufgguf16.0723,162unsloth2026
Qwen3.6-27B-GGUFggufgguf4.0422,088lmstudio-community2026
Qwen_Qwen3.6-27B-GGUFggufgguf2.067,630bartowski2026
Qwen3.6-27Boriginalbf1616.046,778unsloth2026
Qwen3.6-27B-GGUFggufgguf16.020,140ggml-org2026
Qwen-Image-Benchoriginalbf1616.09,764Qwen2026
tmax-27boriginalbf1616.07,955allenai2026
Qwen3.6-27B-UD-MLX-4bitmlxmlx6,982unsloth2026
Qwen3.6-27B-MLX-8bitmlxmlx6,917unsloth2026
Qwen3.6-27B-GGUFggufgguf4.04,137mradermacher2026
Qwen3.6-27B-UD-MLX-NVFP4mlxmlx4,122unsloth2026
Qwen3.6-27B-UD-MLX-6bitmlxmlx2,364unsloth2026
Qwen3.6-27B-UD-MLX-MXFP4mlxmlx1,731unsloth2026
Qwen3.6-27B-i1-GGUFggufgguf1.01,714mradermacher2026
Qwen3.6-27B-MTP-GGUFggufgguf16.01,478ggml-org2026
Qwen3.6-27B-UD-MLX-3bitmlxmlx1,098unsloth2026
Qwen3.6-27B-GGUFggufgguf16.0236Mungert2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.