Qwen3.5-27B
QwenBenchClub cloud
No cloud result yet
Headline numbers come from BenchClub’s pinned cloud runner.
- Arena score
- 1409
- Among open models
- #43of 148
- Among all on Arena
- #144
- Votes
- 28,404
Measured on Qwen3.5-27B, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Chinese1479#22#772K
- Science1436#32#1144.7K
- Software & IT1431#44#14111.2K
- Applied math1430#40#1211.6K
- Math1429#30#1051.7K
- Hard prompts, English1428#43#1358.2K
- Expert1428#38#1252.6K
- Coding1427#44#1447.9K
- English1425#41#13112.6K
- Medicine & health1424#36#1262K
- Hard prompts1414#45#14717.9K
- Long prompts1413#35#12410.9K
- Multi-turn1410#40#1394.9K
- French1409#45#146844
- Spanish1409#42#135926
- Legal & government1405#46#1542.2K
- Business & finance1398#48#1465.6K
- German1395#39#133448
- Instructions1392#43#1469.1K
- Excluding ties1391#44#14620.7K
- Non-English1389#45#15215.8K
- Russian1389#44#1533.3K
- Writing & literature1380#40#1536.4K
- Polish1375#43#155532
- Creative writing1362#41#1564.3K
- Media & sports1362#41#1555.3K
- Korean1357#35#133442
- Japanese1343#34#137225
| Model | Params | Score |
|---|---|---|
| 2.3B | — | |
| 27.8B | 1409 | |
| 36.0B | 1395 | |
| 125B | 1417 | |
| 403B | 1438 |
BenchClub runs
Discussion
- Qwenputer: A functioning, deterministic RISC-V CPU inside Qwen3.5csmantle · Oct 29, 2026709 points
- How to run Qwen 3.5 locallyCuriositry · Mar 7, 2026490 points
- Kev: Tiny Jev-like family of decision models built on top of Qwen3.5tosh · Sep 21, 2026463 points
- Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computerslostmsu · Feb 28, 2026461 points
- Qwen3.5: Towards Native Multimodal Agentsdanielhanchen · Feb 16, 2026434 points
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiounsloth/Qwen3.5-27B-GGUF94 builds
- MLX · Apple silicon: mlx-lm, LM Studiomlx-community/Qwen3.5-27B-mxfp848 builds
- AWQ · vLLM, SGLangQuantTrio/Qwen3.5-27B-AWQ12 builds
- GPTQ · vLLM, SGLangapolo13x/Qwen3.5-27B-quantized.w4a167 builds
Most downloaded
More from Qwen
See more
Qwen3.8-27B6.9M27.8B

Qwen3.5-35B-A3B1.4M36.0B

Qwen3.5-122B-A10B775.1K125B

Qwen3.5-397B-A17B711.9K403B
Other 20–40B models
See more
gemma-4-31B-itGoogle6.3M31.3B
Ornith-1.0-35BOrnith1.4M35.1B
NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1.3M31.6B
GLM-4.7-FlashZ.ai1.2M31.2B