Qwen3.5-122B-A10B
Qwen3.5-122B-A10BAlibabaOpen weightsReasoningProvisionalReleased Mar 4, 2026Input: text, image, video
Overall score
41.7
Rank 133 of 222 · 32.0–51.4
Context window
262K
262,000 tokens
Price per 1M tokens
$0.26 in · $2.08 out
Source: OpenRouter
Blended price
$0.71
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Instruction following#1293.0
- 2Multilingual#987.1
- 3Knowledge#10572.0
- 4Reasoning68.8
- 5Multimodal#3667.8
- 6Coding58.4
- 7Agentic & tool use#10730.4
Benchmark results (17)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Multimodal
| MMMU-Pro | 76.9 | #26 of 48 |
| CharXiv | 77.2 | #37 of 42 |
| MMMU | 83.9 | #6 of 21 |
| MathVision | 86.2 | #13 of 20 |
| V* | 93.2 | #7 of 12 |
| MMVU | 74.7 | #5 of 7 |
Agentic & tool use
| BrowseComp | 63.8 | #37 of 50 |
| Terminal-Bench 2.0 | 49.4 | #31 of 39 |
| OSWorld-Verified | 58.0 | #31 of 35 |
Knowledge
| GPQA | 86.6 | #42 of 117 |
| MMLU-Pro | 86.7 | #10 of 67 |
| SuperGPQA | 67.1 | #11 of 20 |
Coding
| SWE-bench Verified | 72.0 | #56 of 98 |
| LiveCodeBench v6 | 78.9 | #27 of 46 |
Reasoning
| LongBench v2 | 60.2 | #13 of 17 |
Multilingual
| MMLU-ProX | 82.2 | #9 of 16 |
Instruction following
| IFEval | 93.4 | #10 of 53 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price (OpenRouter) | $0.26 | $2.08 | — |
| OpenRouter qwen/qwen3.5-122b-a10b | $0.26 | $2.08 | — |
Open-weight model. Self-hosted or third-party hosted costs vary by provider and infrastructure.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text | 1417 | #123 · 29.7K votes |
| WebDev (coding) | 1361 | #112 · 8.2K votes |
| Vision | 1245 | #70 · 13.2K votes |
Open weights
From the Hugging Face repository.
- Parameters
- 125B
- Licence
- apache-2.0
- Downloads (30 days)
- 702.6K
- Repository
- Qwen/Qwen3.5-122B-A10B ↗
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Similar score
- Gemini 2.5 Flash42.1 · $0.85
- Step 3.5 Flash42.1 · $0.15
- Muse Glimmer 30B41.2 · $0.53
- Claude Haiku 4.541.1 · $2.00
- DeepSeek-R141.1 · $0.96