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 — Qwen3.6 35B A3B
- Cost
- $0.15 in · $1.00 out per 1M tokensBudget
- Memory
- 262K tokens — about 400 pages
- Reads
- Text · Images · Video
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Coding: better than 40% · Agentic work: better than 40% · Overall: better than 33%
- Reliability
- 99.98% uptime, last 24h
- Released
- Apr 27, 2026
Advanced details
- API model ID
qwen/qwen3.6-35b-a3b- Max output
- 236K tokens
- Cached input
- $0.05 per 1M
- Tokenizer
- Qwen
- Coding index
- 41.9
- Agentic work index
- 13.1
- Overall index
- 18.2
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 |
|---|---|---|---|---|---|
| Darkbloom · fp4 | $0.05 | $0.70 | — | 33K | 99.80% |
| AkashML · fp8 | $0.10 | $0.90 | $0.05 | 236K | 99.88% |
| DeepInfra · fp8Degraded | $0.10 | $0.95 | — | 16K | 97.40% |
| Venice · fp8 | $0.10 | $1.00 | — | 66K | 99.67% |
| Reka | $0.15 | $1.00 | $0.05 | 131K | 99.98% |
| Parasail · fp8 | $0.15 | $1.00 | $0.05 | 236K | 99.95% |
| AtlasCloud · fp8 | $0.186 | $1.114 | $0.186 | 66K | 99.16% |
| Phala | $0.20 | $1.27 | — | 236K | 99.67% |
| SiliconFlow · fp8 | $0.24 | $1.80 | $0.15 | 236K | 99.79% |
| CoreWeave · fp8 | $0.25 | $1.25 | $0.25 | 236K | 99.95% |
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