Alibaba
Qwen 3.6 27B
Frontier
5.0MDownloads1.5KCurtidasApr 2026Publicado262K tokensContextoApache 2.0Licença99 ExcepcionalQualidade
Qwen 3.6 27B (27B parameters) requires approximately 18.9 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 22 GB of VRAM.
Comece agora
— copie e cole para rodar localmenteCopy-paste commands to run Qwen 3.6 27B on your machine.
Run
lms load Qwen3.6-27B && lms server startQuick specs
Parameters27B
Architecturedense
Context262K tokens
Modalitytext+vision
Min RAM10.5 GB
Rec. RAM16.5 GB (Q4_K_M)
LicenseApache 2.0
FamilyQwen
✓ Vision✓ Tool Use✓ Thinking✓ Code✓ Chat✓ Reasoning
About this model
- •Dense 27B beats Qwen 3.5 397B-A17B MoE on SWE-bench Verified (77.2%)
- •Fits on RTX 4080/4090/3090 (16-24 GB) at Q4_K_M (~16.8 GB)
- •Multimodal: vision encoder for images, OCR, and hour-scale video
- •262K native context, extensible to ~1M tokens via YaRN
Modelos relacionados
Escolhas rápidas
Melhor hardware
Melhores opções para Qwen 3.6 27B
Rodar este modelo
Opções de quantização
Estimativas de VRAM por nível de quantização
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 10.5 GB | Low | — |
Q3_K_S | 3 | 13.2 GB | Low | — |
NVFP4 | 4 | 15.1 GB | Medium | — |
Q4_K_M | 4 | 16.5 GB | Medium | — |
Q5_K_M | 5 | 19.4 GB | High | — |
Q6_K | 6 | 22.1 GB | High | — |
Q8_0 | 8 | 28.9 GB | Very High | — |
F16 | 16 | 55.4 GB | Maximum | — |
Quality benchmarks
Qwen 3.6 27B benchmark scores
Coding
SWE-bench Verified77.2%
HumanEval+—
Aider Polyglot—
LiveCodeBench83.9%
Reasoning
MMLU-Pro86.2%
GPQA Diamond87.8%
MATH-500—
ARC Challenge—
Source: official · 2026-04-22
Compatibilidade de hardware
Estimativas de compatibilidade para todo o hardware
Computing compatibility...
Detalhamento de memória
Reference: RTX 2060 6GB
Weights16.5 GB
KV Cache1.0 GB
Runtime0.9 GB
Headroom0.6 GB
Perguntas frequentes
FAQ — Qwen 3.6 27B
Veja também