~/model/huggingfaceQwen/Qwen3.6-27B
Alibaba provider

Qwen 3.6 27B

Q2 2026 Qwen3.6 dense flagship. Vision + tool calling; tops the Qwen line on most benchmarks.

Parameters
27B
Family
Alibaba
Context
128K tokens
FP16 weights
54GB
// where you can run it
OllamaLM StudiovLLMMLXoMLX
// hugging face stats (cached daily)
3.9M downloads · 1K likes · license: apache-2.0 · updated 27 days ago

What you need to run this.

$ ./vrambudget --model qwen-3-6-27b --by quant
FP1654 GB
H100 80GB80 GB · fitsM3 Max 9696 GB · fitsRTX Pro 600096 GB · fitsM4 Max 128128 GB · fits
Q8_029 GB
RTX A600048 GB · fitsRTX 6000 Ada48 GB · fitsL40S48 GB · fitsRadeon Pro W790048 GB · fits
Q6_K22 GB
RTX 509032 GB · fitsRTX A600048 GB · fitsRTX 6000 Ada48 GB · fitsL40S48 GB · fits
Q5_K_M19 GB
RTX 309024 GB · fitsRTX 3090 Ti24 GB · fitsRTX 409024 GB · fitsRX 7900 XTX24 GB · fits
Q4_K_M15 GB
RTX 309024 GB · fitsRTX 3090 Ti24 GB · fitsRTX 409024 GB · fitsRX 7900 XTX24 GB · fits

// budgets shown at ctx 8K, concurrency 1, 15% safety headroom. Tune in the calculator →

Alternatives at this size.

$ grep --params similar catalog.json
$ similarGemma 4 26B A4B26BGoogle$ similarMistral Small 324BMistral$ similarQwen3 30B A3B30.5BAlibaba$ similarQwen 2.5 32B32.5BAlibaba

Discussion.

$ gh discussion list

// sign in with github to leave a comment. threads live in the repo's discussions tab.