The Orchestrator › Models › Qwen
Qwen: Qwen3 235B A22B Instruct 2507
Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,...
Specifications
| Provider | Qwen |
| Model ID | qwen/qwen3-235b-a22b-2507 |
| Context window | 262K tokens (262,144) |
| Modalities | text → text |
| Tokenizer | Qwen3 |
| Router sources | OpenRouter, TrustedRouter |
| Released | 2025-07-21 |
| Tracked since | 2026-11-14 |
| Popularity | Not in the ranked set |
| Reasoning | — |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | qwen/qwen3-235b-a22b-2507 | $0.14 | $0.8 | 262K |
| standard | TrustedRouter | qwen/qwen3-235b-a22b-instruct-2507 | $0.095 | $0.58 | 262K |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.095 | $0.58 | 262K |
| 2026-11-14 | $0.14 | $0.8 | 262K |
Capabilities
Agentic tool useStructured outputs256K+ context windowNo content storage (attested)
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, logit_bias, logprobs, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p