Qwen3 235B A22B Instruct 2507 pricing & benchmarks
Qwen3 235B A22B Instruct 2507 is available from Alibaba at $0.090 per million input tokens and $0.550 per million output tokens ($0.205 blended at 3:1). It accepts up to 262,144 tokens of context. 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, inc…
Pricing
Per million tokens · OpenRouter
Input$0.090
Output$0.550
Blended (3:1)$0.205
Specification
qwen/qwen3-235b-a22b-2507
Context window262K
Max output16K
ReleasedJul 2025
LicenceOpen weights
Modalitiesin:text, out:text
No independently-run benchmark scores are published for Qwen3 235B A22B Instruct 2507 yet. We show nothing rather than reprinting an unverified figure.
Where to run Qwen3 235B A22B Instruct 2507
11 options across 10 providers · 2.1× between cheapest and dearest. Blended at 3:1. ⚠ marks an option serving less context than the best available, or re-pricing above a prompt-length threshold. Bracketed labels are the provider's own service tier or region.
| Host | Blended $/1M | In / Out | Context | Weights | Uptime 24h |
|---|---|---|---|---|---|
| DeepInfra | $0.205 | $0.090 / $0.550 | 262K | fp8 | 95.6% |
| Novita ⚠ 131K context, not 262K |
$0.212 | $0.090 / $0.580 | 131K | fp8 | 97.6% |
| Alibaba ⚠ 131K context, not 262K |
$0.262 | $0.150 / $0.598 | 131K | fp8 | 100.0% |
| Venice ⚠ 128K context, not 262K |
$0.300 | $0.150 / $0.750 | 128K | fp8 | 93.7% |
| Nebius | $0.300 | $0.200 / $0.600 | 262K | fp8 | 93.2% |
| Parasail ⚠ 131K context, not 262K |
$0.305 | $0.140 / $0.800 | 131K | fp8 | 99.2% |
| Crusoe | $0.365 | $0.220 / $0.800 | 262K | bf16 | 99.7% |
| StreamLake ⚠ 128K context, not 262K |
$0.367 | $0.210 / $0.840 | 128K | — | 99.6% |
| AtlasCloud ⚠ 131K context, not 262K |
$0.370 | $0.200 / $0.880 | 131K | fp8 | 99.6% |
| Google (us-south1) | $0.385 | $0.220 / $0.880 | 262K | — | 99.9% |
| Google (us-south1) | $0.438 | $0.250 / $1.00 | 262K | — | 99.2% |
Host prices and uptime from OpenRouter, synced 2026-08-08. Lower-precision weights (fp4, fp8) can cost less and score worse than the same model at bf16 — the score above is not host-specific.
Compare Qwen3 235B A22B Instruct 2507
Qwen3 235B A22B Instruct 2507 vs DeepSeek V4 Flash 0731
Qwen3 235B A22B Instruct 2507 vs Claude Opus 5
Qwen3 235B A22B Instruct 2507 vs Gemini 3.6 Flash
Qwen3 235B A22B Instruct 2507 vs Kimi K3
Qwen3 235B A22B Instruct 2507 vs GPT-5.6 Luna
Qwen3 235B A22B Instruct 2507 vs GPT-5.6 Terra
Qwen3 235B A22B Instruct 2507 vs GPT-5.6 Sol
Qwen3 235B A22B Instruct 2507 vs Grok 4.5
FAQ
How much does Qwen3 235B A22B Instruct 2507 cost?
Qwen3 235B A22B Instruct 2507 costs $0.090 per million input tokens and $0.550 per million output tokens on OpenRouter. At a 3:1 input:output mix that blends to $0.205 per million tokens.
What is Qwen3 235B A22B Instruct 2507's context window?
Qwen3 235B A22B Instruct 2507 accepts up to 262,144 tokens of context, and can generate up to 16,384 output tokens.