← All models
Flagship — best answers, highest costOpen Weights
Qwen3 235B A22B Thinking 2507 Pricing
Qwen · qwen-qwen3-235b-a22b-thinking-2507
Input / 1M tokens
$0.230
Output / 1M tokens
$2.30
Context window
131,072 tokens
≈ 175 pages of text
Max output
117,964 tokens
What it costs in practice
Typical request (1,200 in + 400 out tokens)$0.0012
1,000 requests / month$1.20/mo10,000 requests / month$11.96/mo100,000 requests / month$119.60/moEstimate your own workloadOpens the calculator with Qwen3 235B A22B Thinking 2507 preloaded — adjust volume and token counts there.
Price history
List price per 1M tokens since we started tracking in Jun 2026. Each step marks a price change — hover a point for the exact date and rate.
What it excels at
Large parameter count and 256k context support complex multi-step reasoning and long-document analysis. Uniform low token pricing reduces cost for extended outputs.
The business tradeoff
Open-weights deployment requires substantial GPU resources and engineering overhead. Performance consistency depends on the chosen inference stack and quantization.
Cheaper from Qwen
Head-to-head
Vendor list rates, as of Jul 31, 2026 · source: openrouter · per-request examples assume 1,200 input + 400 output tokens.