Qwen3-Coder-480B-A35B-Instruct

Model IntroductionText

Qwen3-Coder-480B-A35B-Instruct: Ultra-large code model, multilingual

Qwen3-Coder-480B-A35B-Instruct is part of the Qwen lineup from Qwen, designed for production workloads that need reliable quality at scale. Ultra-large code model, multilingual.

It accepts Text inputs and produces Coding outputs, with a context window of up to 128,000 input / 8,192 output tokens. The model is tuned for chat, analysis, and tool use in enterprise pipelines.

Strong across code completion, repository-level edits, and multilingual software engineering tasks. Pair with your CI/CD or IDE plugins for automated reviews and patches.

Benchmarks

Benchmark (Area)Score*
LiveCodeBench (Coding)91.6 Pass@1 (Max)
SWE-bench Verified (Coding)79.0% resolved (Max)
HumanEval (Coding)94.2 Pass@1 (Max)
MBPP (Coding)88.5 Pass@1 (Max)
Codeforces Rating (Coding)3052 rating (Max)
Aider Polyglot (Coding)72.4% success (Max)

API request example

Endpoint /v1/chat/completions. Replace <API_KEY> with a key from the console.

Standard OpenAI Chat Completions. Swap `model` for the target slug.

curl https://api.ravenrouter.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <API_KEY>" \
  -d '{
  "model": "Qwen3-Coder-480B-A35B-Instruct",
  "messages": [
    {
      "role": "user",
      "content": "Hello! Introduce yourself in one sentence."
    }
  ],
  "max_tokens": 64
}'

More models: API Reference → Examples by model

Qwen3-Coder-480B-A35B-Instruct API — $0.51/1M input, $1.03/1M output | RavenRouter AI