Route order log in sign up
1 pick your primary route

Qwen: Qwen3 235B A22B Thinking 2507

qwen/qwen3-235b-a22b-thinking-2507 · agent · 128k ctx · added 2025-07-25
$0.4499·3.4998 /M effective hit 21% 48.1M/d

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...

view
openrouter/qwen/qwen3-235b-a22b-thinking-2507
Qwen: Qwen3 235B A22B Thinking 2507 · agent · 128k ctx
48.1M/d 21% $0.4499·3.4998 – 1y
sticker
life 429 tok/d dp age
¢ ¢ᵉ ✓
alibaba
$0.2300·2.3000 · $0.2300·2.3000ⁱ $0.2299·2.2998ᵉ · 0% · – – – · ◐ 443ms · 41t/s · ✕1mo ·
⚡ ¢ⁱ
deepinfra/fp8
$0.2300·2.3000·0.2000 · $0.2237·2.3000ⁱ $0.2299·2.2998ᵉ · 0% · – – – · ✓ 437ms · 53t/s · ✕2mo ·
novita/fp8
$0.3000·3.0000 · $0.3000·3.0000ⁱ $0.2999·3.0000ᵉ · 0% · – – – · ✓ 1478ms · 24t/s · ✕1mo ·
♻ ★
venice/fp8
$0.4500·3.5000 $0.4500·3.5000ⁱ $0.4499·3.4998ᵉ 21% – – 100% ✓ 902ms 33t/s 3mo ·
requesty/alibaba/qwen3-235b-a22b-thinking-2507
qwen3-235b-a22b-thinking-2507 · agent · 262k ctx
– $0.2237·2.3000 – 1y
hf/Qwen/Qwen3-235B-A22B-Thinking-2507
Qwen3-235B-A22B-Thinking-2507 · agent · 262k ctx
– $0.2300·2.3000 – 1y
compare all models →