Model Library
AlibabaReleased February 24, 2026Qwen3.5 122B A10B
Alibaba · Released February 24, 2026
Intelligence score
32
At a glance
- Intelligence
- 32
- Speed
- 139.4tokens/sec
- Latency
- 2.62stime to first token
- Input price
- $0.40per 1M tokens
- Output price
- $3.20per 1M tokens
- Context window
- 260Ktokens
- Parameters
- 125billions
Summary
Comparison summary
Qwen3.5 122B A10B (Reasoning) scores 32 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9). Qwen3.5 122B A10B (Reasoning) generates output at 139.4 tokens per second (based on Alibaba's API), which is above average compared to other open weight models of similar size (median: 88.2 t/s). Qwen3.5 122B A10B (Reasoning) costs $0.40 per 1M input tokens (better than average, median: $0.40) and $3.20 per 1M output tokens (at the higher end, median: $0.84), based on Alibaba's API.
Specifications
Technical specifications
| Model name | Qwen3.5 122B A10B |
|---|---|
| Provider | Alibaba |
| Release date | February 24, 2026 |
| Intelligence score | 32 |
| Speed | 139.4 tokens/sec |
| Latency (TTFT) | 2.62s |
| Input price | $0.40 / 1M tokens |
| Output price | $3.20 / 1M tokens |
| Context window | 260,000 tokens |
| Parameters | 125B |
| Reasoning | Yes |
| Input modalities | text, image |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: qwen3-5-122b-a10b |
Pricing
Token prices
Price per 1M tokens
Qwen3.5 122B A10B, as listed in the catalog record
- Input$0.40
- Output$3.20
| Token type | Price per 1M |
|---|---|
| Input | $0.40 |
| Output | $3.20 |
Model Intelligence legacy profile: qwen3-5-122b-a10b
FAQ · 21
Frequently asked questions
Qwen3.5 122B A10B (Reasoning) was released on February 24, 2026.
Qwen3.5 122B A10B (Reasoning) was created by Alibaba.
Qwen3.5 122B A10B (Reasoning) scores 32 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9).
Qwen3.5 122B A10B (Reasoning) generates output at 139.4 tokens per second (based on Alibaba's API), which is above average compared to other open weight models of similar size (median: 88.2 t/s).
Qwen3.5 122B A10B (Reasoning) has a time to first token (TTFT) of 2.62s (based on Alibaba's API), which is at the higher end compared to other open weight models of similar size (median: 1.52s).
Qwen3.5 122B A10B (Reasoning) costs $0.40 per 1M input tokens (better than average, median: $0.40) and $3.20 per 1M output tokens (at the higher end, median: $0.84), based on Alibaba's API.
Qwen3.5 122B A10B (Reasoning) costs $0.40 per 1M input tokens and $3.20 per 1M output tokens (based on Alibaba's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.68 per 1M tokens. Pricing may vary by provider.
When evaluated on the Intelligence Index, Qwen3.5 122B A10B (Reasoning) generated 96M output tokens, which is somewhat higher than average compared to other open weight models of similar size (median: 59M).
Yes, Qwen3.5 122B A10B (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
Qwen3.5 122B A10B (Reasoning) supports text and image input.
Qwen3.5 122B A10B (Reasoning) supports text output.
Yes, Qwen3.5 122B A10B (Reasoning) supports image input and can analyze, describe, and answer questions about images.
Yes, Qwen3.5 122B A10B (Reasoning) is multimodal. It can process text and image input and generate text output.
Qwen3.5 122B A10B (Reasoning) has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Qwen3.5 122B A10B (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Qwen3.5 122B A10B (Reasoning) has 125 billion parameters (10 billion active).
Qwen3.5 122B A10B (Reasoning) is a Mixture of Experts (MoE) model with 125 billion total parameters, but only 10 billion active parameters are used during inference.
Qwen3.5 122B A10B (Reasoning) is released under the Apache 2.0 license. This license allows commercial use.
Qwen3.5 122B A10B (Reasoning) achieves a score of 32 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Qwen3.5 122B A10B (Reasoning) is available via API through 5 providers.
Qwen3.5 122B A10B (Reasoning) is available through 5 API providers.