Qwen: Qwen3.5-122B-A10B
Analysis Summary
Qwen3.5-122B-A10B is a 122-billion-parameter mixture-of-experts model from Alibaba's Qwen team, activating 10B parameters per pass. Its intelligence index places it in the 'very strong' tier, closely matching the 27B dense sibling, with comparable agentic and coding scores. Vision, tool use, and function calling are all supported, and instruction-following is reliable across a 262K context window.
For businesses, the model suits similar workloads to the 27B variant: agentic automation, long-document analysis, structured content, and multimodal tasks. The MoE architecture means inference cost is lower than a dense 122B model would imply, though the 27B dense model offers nearly identical benchmark performance at slightly lower input cost. Coding capability is present but not a primary strength.
At $0.26/1M input and $2.08/1M output, it is priced reasonably for its capability class. A -4 point regional accessibility adjustment applies. Teams that need the larger parameter count for specific tasks or prefer the MoE architecture will find it a capable option, though the 27B model may be more cost-efficient for most workloads.
Assessed June 6, 2026
Editorial notes
Qwen3.5-122B-A10B is a large sparse MoE model with very strong reasoning, a 62.3 agentic index, vision, tool use, and function calling at $0.26/1M input across a 262K context window.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Qwen: Qwen3.5-122B-A10B compares
Qwen: Qwen3.5-122B-A10B ranks #54 of 377 AI models we track for overall intelligence, #70 of 314 for coding, #47 of 289 for agentic tasks. Its 262K-token context window is larger than 81% of the models we list. At $0.26 per million input tokens it is cheaper than 45% of comparable models.
About Qwen: Qwen3.5-122B-A10B
The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Qwen: Qwen3.5-122B-A10B stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
qwen/qwen3.5-122b-a10b
|
| Provider | qwen |
| Release Date | February 25, 2026 |
| Context Length | 262,144 tokens |
| Max Completion | 262,144 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.26 | $0.000260 |
| Output | $2.08 | $0.002080 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Qwen: Qwen3.5-122B-A10B
How much does Qwen: Qwen3.5-122B-A10B cost?
Qwen: Qwen3.5-122B-A10B costs $0.26 per million input tokens and $2.08 per million output tokens.
What is the context window of Qwen: Qwen3.5-122B-A10B?
Qwen: Qwen3.5-122B-A10B has a context window of 262,144 tokens (262K).
Is Qwen: Qwen3.5-122B-A10B good for coding?
On our coding benchmark index, Qwen: Qwen3.5-122B-A10B ranks #70 of 314 models, placing it in the top quartile of the field for code generation and debugging.
What can Qwen: Qwen3.5-122B-A10B do?
Qwen: Qwen3.5-122B-A10B supports image/vision input, tool use, and function calling.
Who created Qwen: Qwen3.5-122B-A10B?
Qwen: Qwen3.5-122B-A10B is developed by Qwen and was released on February 25, 2026.
