Qwen3 Coder 30B A3B Instruct
Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the Qwen3 architecture, it supports a native context length of 256K tokens (extendable to 1M with Yarn) and performs strongly in tasks involving function calls, browser use, and structured code completion. This model is optimized for instruction-following without “thinking mode”, and integrates well with OpenAI-compatible tool-use formats.
- Input / output price
- $0.15 / $0.60 per 1M
- Context
- 262K tokens
- Released
- Jul 31, 2025
Paste into Claude Code, Cursor, or Codex
Help me set up the Respan gateway: fetch and follow https:/
Providers
3 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.
| Provider | Input | Output |
|---|---|---|
Bedrock bedrock/qwen3-coder-30b-a3b-instruct | $0.15/M | $0.60/M |
Concentrate concentrate/qwen3-coder-30b-a3b | $0.15/M | $0.60/M |
OpenRouter openrouter/qwen3-coder-30b-a3b-instruct | $0.15/M | $0.60/M |
Is a provider down? AWS Bedrock status, OpenRouter status
Pricing
Averaged across 3 providers, in USD per 1M tokens after gateway discounts.
- Average input
- $0.15 per 1M
- Average output
- $0.60 per 1M
Cost by volume
Total cost from 0 to 1T tokens.
Throughput and latency
30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.
Output speed
Higher is better21tokens/sacross providers
Response latency
Lower is better2.1sacross providers
Benchmarks
13 published scores from 1 source, each linked to where it was published.
Published scores
| Benchmark | Score | Source | Published |
|---|---|---|---|
| SWE-bench Verified (w/ OpenHands, 500 turns)huggingface.co | 51.6% | huggingface.co | Jul 31, 2025 |
| SWE-bench Multilingualhuggingface.co | 34.7% | huggingface.co | Jul 31, 2025 |
| Aider-Polyglothuggingface.co | 33.3% | huggingface.co | Jul 31, 2025 |
| Terminal-Benchhuggingface.co | 31.3% | huggingface.co | Jul 31, 2025 |
| SWE-bench Livehuggingface.co | 20.7% | huggingface.co | Jul 31, 2025 |
| Multi-SWE-bench minihuggingface.co | 19.5% | huggingface.co | Jul 31, 2025 |
| Multi-SWE-bench flashhuggingface.co | 19.3% | huggingface.co | Jul 31, 2025 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| BFCL-v3huggingface.co | 62.2% | huggingface.co | Jul 31, 2025 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| TAU-Bench Retailhuggingface.co | 68.7% | huggingface.co | Jul 31, 2025 |
| Mind2Webhuggingface.co | 51% | huggingface.co | Jul 31, 2025 |
| TAU-Bench Airlinehuggingface.co | 48% | huggingface.co | Jul 31, 2025 |
| WebArenahuggingface.co | 43.5% | huggingface.co | Jul 31, 2025 |
| Spider2huggingface.co | 21.4% | huggingface.co | Jul 31, 2025 |
Apps
Use Qwen3 Coder 30B A3B Instruct in the tools you already run. Each guide points the app at the Respan gateway.
Setup guides for more apps are in the integration docs.
Qwen3 Coder 30B A3B Instruct FAQ
Short answers, built from the data on this page.
Qwen3 Coder 30B A3B Instruct is a model from Qwen, released on Jul 31, 2025. Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the Qwen3 architecture, it supports a native context length of 256K tokens (extendable to 1M with Yarn) and performs strongly in tasks involving function calls, browser use, and structured code completion. This model is optimized for instruction-following without “thinking mode”, and integrates well with OpenAI-compatible tool-use formats.
On the Respan gateway, Qwen3 Coder 30B A3B Instruct costs $0.15 per 1M input tokens and $0.60 per 1M output tokens on every provider, after gateway discounts.
Qwen3 Coder 30B A3B Instruct is available through 3 providers: Bedrock, Concentrate, and OpenRouter.
Qwen3 Coder 30B A3B Instruct accepts up to 262K tokens of context on its largest provider route.
Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "qwen3-coder-30b-a3b-instruct". To set it up with a coding agent, copy the agent setup prompt at the top of this page.
Qwen3 Coder 30B A3B Instruct has 13 published benchmark scores, including SWE-bench Verified (w/ OpenHands, 500 turns) 51.6%, BFCL-v3 62.2%, and TAU-Bench Retail 68.7%. Each score links to where it was published; the data is collected by Respan.
Certified where it counts.
Built to meet the security and privacy standards that enterprise and healthcare teams require.
ISO 27001
The internationally recognized standard for information security management.
SOC 2
Secure, compliant management of your data across all of our systems.
GDPR
Operated under GDPR, the world's strictest standard for data privacy.
HIPAA
HIPAA compliant, with a BAA available for healthcare teams.