Qwen3.5-397B-A17B

AlibabaOpenApache 2.0

Qwen3.5-397B-A17B is a vision-language model from Alibaba. It ranks 54 of 167 on LMArena’s overall vision board with 1262±5 from 33,611 votes. Its best category is Entity recognition (#14 of 52). License: Apache 2.0.

Key figures

LMArena · Overall1262±5#54 / 167 · Votes 33,611
OCR1270±5#53 / 123 · Votes 23,370
Diagrams1277±7#51 / 123 · Votes 8,845
Homework1292±9#43 / 118 · Votes 4,164
Captioning1227±57#15 / 34 · Votes 78
Entity recognition1271±37#14 / 52 · Votes 226
Creative writing1262±13#46 / 106 · Votes 2,414
Humor1264±17#47 / 102 · Votes 1,260
English1269±7#49 / 167 · Votes 12,989
Chinese1315±15#48 / 135 · Votes 2,068
Developer
Alibaba
License
Apache 2.0
First seen (our snapshots)
2026-10-26

Who it suits

  • Overall #54 of 167: upper half.
  • Strongest category: Entity recognition (#14 of 52).
  • Open weights (Apache 2.0): you can self-host it.

Neighbours on the overall board

#ModelScoreVotesLicenseΔ vs this model
51Gemini 2.5 Pro Google126388,553Proprietary+1
52SStep 5 Preview StepFun12633,523Proprietary+1
53Grok 4.20 Beta (Reasoning, 0309) xAI126326,586Proprietary+1
54Qwen3.5-397B-A17B Alibaba126233,611Apache 2.0
55Gemma 4 26B A4B Google126023,747Apache 2.0-2
56MiMo-V2.6-Flash Xiaomi12606,906MIT-2
57Grok 4.20 Multi-Agent Beta (0309) xAI125924,357Proprietary-3

Alternatives · More from Alibaba

#ModelScoreVotesLicense
11Qwen3.8-Max Alibaba13129,842Proprietary
39Qwen3.7-Plus Alibaba127715,152Proprietary
42Qwen3.8-27B Alibaba12709,632Apache 2.0
66Qwen3-VL-235B-A22B-Instruct Alibaba124712,672Apache 2.0
69Qwen3.5-122B-A10B Alibaba124413,171Apache 2.0

Alternatives · Top open-weight models

#ModelScoreVotesLicense
24GLM-5.3-Flash Z.ai129531,530MIT
34Kimi K2.6 Moonshot AI128216,265Modified MIT
36Gemma 4 31B Google127941,787Apache 2.0
37DeepSeek-V4.1-Flash-20260910 (Max) DeepSeek12794,247MIT
42Qwen3.8-27B Alibaba12709,632Apache 2.0

How to read these scores

  • If two scores differ by less than their ± intervals, the models are statistically close.
  • Each category board is computed independently; do not compare scores across categories.
  • Models with few votes move more; ranks of new entries may change.

FAQ

Where does Qwen3.5-397B-A17B rank among vision models?

Overall rank 54 (1262); OCR rank 53 (1270); Diagrams rank 51 (1277); Homework rank 43 (1292); Captioning rank 15 (1227); Entity recognition rank 14 (1271); Creative writing rank 46 (1262); Humor rank 47 (1264); English rank 49 (1269); Chinese rank 48 (1315) (LMArena, data as of Nov 14, 2026).

How good is Qwen3.5-397B-A17B at OCR?

Qwen3.5-397B-A17B ranks 53 of 123 on LMArena’s OCR board with 1270 (23,370 votes).

Is Qwen3.5-397B-A17B better with Chinese or English prompts?

Chinese board: #48 of 135 (1315); English board: #49 of 167 (1269). The boards are computed separately, so compare ranks rather than scores.

Is Qwen3.5-397B-A17B open-weight?

LMArena lists its license as Apache 2.0, not proprietary; check the license text for commercial use.

How far is Qwen3.5-397B-A17B from #1?

#1 overall is Claude Fable 5 (High) (1326); Qwen3.5-397B-A17B has 1262, 64 points behind.

Sources

LMArena data published Nov 9, 2026.

Synced here Nov 14, 2026.

Price, context length and release date have no reliable public data here and are not shown; all figures come from snapshots, blanks are –.