> Qwen3.5 is an open-weights AI model from Alibaba, released 16 Feb 2026. Index score 146 (GPT-4 = 100), rank 68 of 229 ranked models. API price $0.55 in / $3.5 out per million tokens, 262K context. [Models](https://themodelindex.org/models/) /[Alibaba](https://themodelindex.org/labs/alibaba/) # Qwen3.5 Alibaba Released 16 Feb 2026 Open [Compare](https://themodelindex.org/compare/qwen3-5-vs-gpt-5-4/) [Run it yourself →](https://themodelindex.org/local/) [Announcement](https://qwenlm.github.io/qwen-code-docs/en/blog/updates/weekly-update-2026-03-03/) 146134–158 Index score #68 of 229 models 3/ 8 Core tests taken 3 of 6 areas · plus 5 anchor tests −24 Versus the frontier at release best then: GPT-5.3-Codex $0.55/ $3.5 Price per million tokens input / output, list price 262K Context window tokens it reads at once ## Where it sits · on the frontier _Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_ o3 reached this score 10 months earlier. ## Test results · by area · tick = best by any model Show 5 anchor tests Reasoning & knowledge evidence GPQA Diamond 86% Humanity's Last Exam no result Maths evidence FrontierMath T1-3 31% Coding no evidence yet SWE-bench Verified no result Terminal-Bench no result Agents evidence APEX-Agents 25% Novel problems no evidence yet ARC-AGI-2 no result Long tasks no evidence yet METR time horizon no result Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above. ## What others say · rank on each leaderboard The Model Index independent tests, our method #68 of 219 · 146 LMArena · Creative writing creative writing #43 of 386 · 1446 LMArena · Coding coding questions in chat #44 of 383 · 1514 LMArena · Text overall chat quality, judged by people #48 of 388 · 1465 LMArena · Maths maths questions in chat #48 of 373 · 1468 Artificial Analysis Intelligence Index overall capability across ten evaluations #53 of 207 · 21 Epoch Capabilities Index overall capability from many benchmarks #67 of 239 · 146.7 LMArena · Vision understanding images #48 of 149 · 1245 LMArena · WebDev building web apps and front ends #77 of 123 · 1400 Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source. ## Best at · top three among current models Not in the top three among current models on any category we track. Where this model places in the top three of today's models, on category leaderboards and on our core tests. ## Where to use it · prices checked 1 Oct 2026 #### From the lab Qwen Chat app · subscription ↗ Alibaba Cloud Model Studio API ↗ #### API providers · 10 · per million tokens, in / out A Alibaba $0.39 / $2.34 cheapest D DeepInfra $0.45 / $3.0 · fp8 ↗ P Parasail $0.50 / $3.6 · fp8 ↗ D DigitalOcean $0.55 / $3.5 ↗ P Phala $0.55 / $3.5 ↗ A AtlasCloud $0.55 / $3.5 · fp8 ↗ S StreamLake $0.60 / $3.6 ↗ G GMICloud $0.60 / $3.6 · fp8 ↗ #### Run it yourself 🤗 Official weights Qwen/Qwen3.5-397B-A17B ↗ Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps. ## Facts and sources Released 16 Feb 2026 Announcement [qwenlm.github.io ↗](https://qwenlm.github.io/qwen-code-docs/en/blog/updates/weekly-update-2026-03-03/) Release entry Qwen3.5 (397B open + 3.5-Plus) Scored as Qwen3.5 397B-A17B Artificial Analysis index 18 Output speed 85 tokens/s Hugging Face [Qwen/Qwen3.5-397B-A17B ↗](https://huggingface.co/Qwen/Qwen3.5-397B-A17B) ## Alibaba releases · around this one Qwen3-Next-80B-A3B Open 11 SEP 2025 Qwen3-Max-Thinking Closed 26 JAN 2026 Qwen3.5 Open 16 FEB 2026 index 146 Qwen3.6-Plus Closed 2 APR 2026 index 148 Qwen3.6-35B-A3B Open 16 APR 2026 index 139 --- Source: https://themodelindex.org/models/qwen3-5/ · The Model Index · data as of 2026-10-01