Qwen3 235B A22B Thinking 2507

ChatInput: TextReleased Jul 25, 2025

Built by Alibaba (Qwen) · China · qwen.ai

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...

How to use

model: "qwen/qwen3-235b-a22b-thinking-2507"

Cap quality here and save

Hard questions go to this model, easy ones drop to something cheaper. Set a savings target (0–0.8) and the router tunes itself until realized savings converge on it.

{
  "model": "celon/auto",
  "celon": {
    "anchor": "qwen/qwen3-235b-a22b-thinking-2507",
    "savings_target": 0.4
  }
}

Naming a model directly gets you that model — savings controls don't apply.

Context

262K

Max output

—

Input $/1M

$0.23

Output $/1M

$2.30

Features

Tool callingJSON

Serving providers

DeepInfraNovitaW&B Inference

Providers

Price, latency, and uptime per provider serving this model. Celon routes each request to the fastest and cheapest of them. Click a row for regions and data policies.

Loading provider metrics…

Performance

Benchmark quality scores, and measured speed per provider.

Quality

Epoch Capabilities IndexComposite over 50+ benchmarks

144.8

GPQA DiamondPhD-level science questions

80.0%

AIMECompetition mathematics

86.7%

SimpleQA VerifiedFactual accuracy

50.1%

At each model's best reasoning effort. Benchmarks it wasn't run on are omitted.

Speed

Loading provider metrics…