Qwen3-Coder: API Pricing, Context Window & Benchmarks
Qwen3-Coder is a language model from Qwen, released in January 2025.
Qwen3-Coder is an advanced coding model with 480B parameters and 35B active parameters. This is a convenience alias for qwen3-coder-480b-a35b-instruct. It excels at code generation, understanding, and software development tasks with a
Qwen3-Coder pricing
Providers
Qwen3-Coder starts at $0.180 per million input tokens and $0.180 per million output tokens via DeepInfra. See all 2 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 DeepInfra | $0.180 | — | $0.180 | 256.0K/256.0K | — | — | / |
 Fireworks | $0.250 | — | $0.250 | 256.0K/256.0K | — | — | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
Qwen3-Coder model size
Qwen3-Coder has 480 billion parameters. See how it compares to other models in the same parameter range.
Qwen3-Coder context window
Input and output token limits for Qwen3-Coder, plus how it ranks on long-context understanding.
Qwen3-Coder latency
Qwen3-Coder time to first token, sustained output throughput, and failed-request rate from live API traffic over the trailing 7 days.
Qwen3-Coder examples
Recent arena outputs from Qwen3-Coder, picked from the highest-ranked matchups.
Qwen3-Coder license
Qwen3-Coder is released under the Apache 2.0 license, which permits commercial use, has 480.0B parameters.
- License
- Apache 2.0
- Commercial use allowed
- Parameters
- 480.0B
Apache License 2.0 - allows commercial use
Qwen3-Coder resources
Official sources for Qwen3-Coder: official playground, official launch post, source repository, model weights.
FAQ
Common questions about Qwen3-Coder.