Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Kimi K3
GLM-5.3
GPT-5.5
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 Preview
Qwen3.8-Max
Qwen3.8-Max is a proprietary model from Alibaba, released Sep 3, 2026. It ranks #15 of 351 on the LMArena text leaderboard with a rating of 1484, and #11 of 27 on the Vals Index with 59.05%. It costs $2 per million input tokens and $6 per million output tokens, with a 1M-token context window.
- Developer
- Alibaba Qwen
- Released
- Sep 3, 2026
- Context window
- 1M
- Price in / out, per 1M tokens
- $2 / $6
- Weights
- Proprietary
- Inputs
- Text, Images, Video
- Reasoning
- Yes
- Tool use
- Yes
The rating LMArena published at each update since Aug 2, 2026. Ratings move as votes come in and new models join.
Benchmarks
| Benchmark | Score bar | Accuracy | Rank | Cost / test |
|---|---|---|---|---|
| Vals IndexVals AI | 59.05% | 11 / 27 | $8.73 | |
| Terminal-Bench 4.0Coding | 34.34% | 18 / 55 | $10.63 | |
| SWE-benchCoding | 85.60% | 16 / 85 | $1.13 | |
| Vibe Code BenchCoding | 64.70% | 49 / 115 | $8.24 | |
| Code MigrationCoding | 57.19% | 10 / 28 | $27.08 | |
| LiveCodeBenchCoding | 87.85% | 10 / 143 | $2 / $6 | |
| Finance AgentFinance | 58.83% | 23 / 31 | $2.99 | |
| Finance IndexFinance | 61.66% | 5 / 31 | $3.83 | |
| Tax Agent BenchFinance | 32.09% | 22 / 76 | $1.55 | |
| Legal IndexLaw | 73.90% | 5 / 28 | $4.25 | |
| Legal Research BenchLaw | 46.15% | 17 / 85 | $5.31 | |
| MedCodeMedicine | 40.67% | 62 / 107 | $2 / $6 | |
| GPQA DiamondReasoning | 93.69% | 6 / 138 | $2 / $6 | |
| MMLU-ProReasoning | 88.60% | 18 / 138 | $2 / $6 | |
| MMMUReasoning | 88.03% | 12 / 93 | $2 / $6 | |
| Terminal-Bench ScienceReasoning | 12.86% | 13 / 42 | $16.68 |
Vals AI, as of Oct 24, 2026.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1484 | 15 / 351 | 30.5K |
| CodingLMArena | 1503 | 21 / 346 | 8.3K |
| MathLMArena | 1501 | 15 / 339 | 1.5K |
| Creative writingLMArena | 1477 | 16 / 349 | 6.1K |
| Instruction followingLMArena | 1481 | 18 / 351 | 10.8K |
| Hard promptsLMArena | 1497 | 16 / 351 | 20.1K |
| Multi-turnLMArena | 1490 | 14 / 349 | 4.6K |
| Long queriesLMArena | 1489 | 18 / 330 | 13.5K |
| ExpertLMArena | 1507 | 23 / 305 | 3.3K |
| Legal and governmentIndustries | 1494 | 16 / 325 | 2.3K |
| Medicine and healthcareIndustries | 1503 | 9 / 325 | 2K |
| Business and financeIndustries | 1481 | 12 / 344 | 5.6K |
| ChineseLanguages | 1539 | 12 / 328 | 2.3K |
| SpanishLanguages | 1487 | 12 / 249 | 973 |
| GermanLanguages | 1476 | 24 / 261 | 478 |
| FrenchLanguages | 1502 | 11 / 252 | 1.1K |
| JapaneseLanguages | 1473 | 22 / 236 | 550 |
| WebDevLMArena | 1673 | 12 / 124 | 15.6K |
| VisionLMArena | 1313 | 10 / 146 | 10K |
LMArena, CC BY 4.0, as of Oct 22, 2026. Where a model is rated at several effort levels, its best is shown (Qwen3.8-Max, Qwen3.8-Max (0902)).
Compare Qwen3.8-Max
Alibaba Qwen
Alibaba Cloud’s Qwen team releases a broad family of open-weight models, from small to frontier scale. Founded 2023 (Qwen), based in Hangzhou. qwenlm.github.io
Questions
Who makes Qwen3.8-Max?
Qwen3.8-Max is made by Alibaba Qwen, founded 2023 (Qwen) and based in Hangzhou.
How much does Qwen3.8-Max cost?
$2 per million input tokens and $6 per million output tokens, at its provider’s list price.
What is Qwen3.8-Max’s context window?
1M tokens (1,000,000).
How does Qwen3.8-Max rank?
#15 of 351 on LMArena’s text leaderboard, rated 1484 (95% interval 1479–1488) from 30,500 votes, as of Oct 22, 2026.
Is Qwen3.8-Max open source?
No. Qwen3.8-Max is proprietary and served through its provider’s API.