Qwen
Alibaba's open-weight family, with a coder model in nearly every size.
Qwen is Alibaba's model family and the largest open-weight line there is — dozens of sizes from a few billion parameters up to the Max models, most of them published so you can run them yourself. For vibe coders the draw is the Qwen Coder models: strong at code for the price, cheap to call through a provider, and small enough at the low end to run on your own machine. The flagship Max models take text, images and video and go up to a million tokens of context.
At a glance — Qwen3.5 397B A17B
- Cost
- $0.55 in · $3.50 out per 1M tokensMid-range
- Memory
- 262K tokens — about 400 pages
- Reads
- Text · Images · Video
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Coding: better than 49% · Agentic work: better than 40% · Overall: better than 34%
- Reliability
- 99.92% uptime, last 24h
- Released
- Feb 16, 2026
Advanced details
- API model ID
qwen/qwen3.5-397b-a17b- Max output
- 236K tokens
- Max input
- 258K tokens
- Cached input
- $0.225 per 1M
- Tokenizer
- Qwen3
- Coding index
- 48.2
- Agentic work index
- 8.3
- Overall index
- 18.4
Providers
The same model, hosted by different companies. OpenRouter picks one per request and falls back to the next if it fails.
| Provider | Input /1M | Output /1M | Cached /1M | Max output | Uptime 24h |
|---|---|---|---|---|---|
| Alibaba | $0.39 | $2.34 | — | 66K | 99.84% |
| DeepInfra · fp8 | $0.45 | $3.00 | $0.22 | 82K | 99.65% |
| Parasail · fp8 | $0.50 | $3.60 | $0.30 | 236K | 99.92% |
| DigitalOcean | $0.55 | $3.50 | $0.11 | 118K | 92.64% |
| Phala | $0.55 | $3.50 | $0.225 | 236K | 95.10% |
| AtlasCloud · fp8 | $0.55 | $3.50 | $0.55 | 66K | 94.39% |
| StreamLake | $0.60 | $3.60 | $0.12 | 64K | 97.84% |
| GMICloud · fp8 | $0.60 | $3.60 | — | 236K | 86.32% |
| Novita | $0.60 | $3.60 | — | 66K | 97.98% |
| Venice | $0.75 | $4.50 | — | 33K | 86.19% |
Supported parameters
- frequency_penalty
- include_reasoning
- logit_bias
- logprobs
- max_tokens
- min_p
- presence_penalty
- reasoning
- repetition_penalty
- response_format
- seed
- stop
- structured_outputs
- temperature
- tool_choice
- tools
- top_k
- top_logprobs
- top_p
Live from OpenRouter, refreshed hourly. Scores from Artificial Analysis.
Key info
- Pricing
- Open source
- Category
- Models
- Best for
- Intermediate
- Made by
- Alibaba
- Weights
- Open, most sizes
- Good for
- Cheap coding models, and running one yourself