Pass IndexThe State of AISign in

Qwen3 235B A22B Thinking 2507

Qwen3-235B-A22B-Thinking-2507 is the Qwen3's new model with scaling the thinking capability of Qwen3-235B-A22B, improving both the quality and depth of reasoning.

text → text · made by Alibaba

$0.13$0.6per Mtok in / out
SiliconFlow · 12 sellers

Sold by 17 ways

SellerLaneRate
SiliconFlow aggregatorstandard$0.13per Mtok in$0.6per Mtok outmodels.dev · read 2026-09-16
Requesty aggregatornot on the seller's list todaydeepinfra flex$0.17per Mtok in$1.66per Mtok out$0.14per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatordeepinfra$0.18per Mtok in$1.84per Mtok out$0.16per Mtok cachedrouter.requesty.ai · read 2026-09-14
Fireworks AI aggregatorstandard$0.22per Mtok in$0.88per Mtok outraw.githubusercontent.com · read 2026-09-16
DeepInfra aggregatorstandard$0.23per Mtok in$2.3per Mtok outapi.deepinfra.com · read 2026-08-25
Hugging Face Inference Providers 3 ways$0.3per Mtok in$3per Mtok out
Hugging Face Inference Providers aggregatordeepinfra$0.23per Mtok in$2.3per Mtok outrouter.huggingface.co · read 2026-08-25
Hugging Face Inference Providers aggregatorstandard$0.3per Mtok in$3per Mtok outmodels.dev · read 2026-09-16
Hugging Face Inference Providers aggregatornovita$0.3per Mtok in$3per Mtok outrouter.huggingface.co · read 2026-08-25
Nous Research aggregatorstandard$0.23per Mtok in$2.3per Mtok outinference-api.nousresearch.com · read 2026-09-12
OpenRouter 3 ways$0.23per Mtok in$2.3per Mtok out
OpenRouter aggregatorstandard$0.23per Mtok in$2.3per Mtok outopenrouter.ai · read 2026-08-24
OpenRouter aggregatornovita/fp8$0.3per Mtok in$3per Mtok outopenrouter.ai · read 2026-08-26
OpenRouter aggregatorvenice/fp8$0.45per Mtok in$3.5per Mtok outopenrouter.ai · read 2026-08-25
ElectronHub aggregatorstandard$0.24per Mtok in$0.72per Mtok outapi.electronhub.ai · read 2026-09-16
Novita AI aggregatorstandard$0.3per Mtok in$3per Mtok outapi.novita.ai · read 2026-08-25
Venice AI aggregatorstandard$0.45per Mtok in$3.5per Mtok outmodels.dev · read 2026-09-16
Together AI aggregatorstandard$0.65per Mtok in$3per Mtok outraw.githubusercontent.com · read 2026-09-16
PPIO aggregatorstandard$2per Mtok in$20per Mtok outapi.ppinfra.com · read 2026-09-16

Measured 24 standings

PlaceBoardMetricScore
11thof 42Fictionlivebench — Epoch AI120k token score0.688
13thof 49Lech mazur writing — Epoch AIMean score8.24
40thof 80Simpleqa verified — Epoch AImean_score40.44
57thof 60Vending bench 2 — Epoch AIScore-11.34
59thof 101FrontierMath (Tiers 1-3)mean_score8.481
64thof 72Frontiermath tier 4 — Epoch AImean_score0
66thof 161Dtbench — Epoch AIAccuracy80.27
67thof 123Lmca — Epoch AIScore29.31
80thof 290OTIS Mock AIME 2024-2025mean_score86.667
103rdof 127Mystery game puzzles — Epoch AImean_score9
105thof 168Scicode — Epoch AIScore42.361
105thof 170Weirdml — Epoch AIAccuracy41.04
122ndof 313GPQA Diamondmean_score80.051
125thof 143Artificial Analysis · Agentic IndexIndex1.3
126thof 147Design Arena · 3d (models)Elo1026
129thof 222Chess puzzles — Epoch AImean_score12
133rdof 156Design Arena · codecategories (models)Elo1048
134thof 162Design Arena · website (models)Elo1065
140thof 150Design Arena · uicomponent (models)Elo960
143rdof 152Design Arena · dataviz (models)Elo964
144thof 184Artificial Analysis · Coding IndexIndex22.1
146thof 155Design Arena · gamedev (models)Elo980
165thof 177Critpt — Epoch AIAccuracy0
356thof 656Epoch capabilities index — Epoch AIECI Score143.88

About

Qwen3 235B A22B Thinking 2507 — a text model from Alibaba, sold by 12 companies from $0.13 in and $0.6 out per million tokens, placed 11th of 42 on Fictionlivebench — Epoch AI.

It takes text and returns text, with a context window of 131,072 tokens. It was published in April 2025, trained on material up to April 2025. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning and agents. Twelve companies sell it. The cheapest is $0.13 in and $0.6 out per million tokens at SiliconFlow. Beside the standard rate there are separately routed and regional lanes. It has been measured on 24 boards, and stands best at 11th of 42 on Fictionlivebench — Epoch AI.

Every current figure

Maker
Alibaba
Register
model
Takes
text
Returns
text
Context
131,072 tokens
Published
April 2025
Knowledge to
April 2025
Parameters
235 billion
Licence
not read
Sellers
12
Maker's own price
not read
Price
$0.13 in and $0.6 out per million tokens — SiliconFlow
Boards
24
Best place
11th of 42 — Fictionlivebench — Epoch AI

Known as 3 names

Qwen/Qwen3-235B-A22B-Thinking-2507deepinfra/Qwen/Qwen3-235B-A22B-Thinking-2507deepinfra/Qwen/Qwen3-235B-A22B-Thinking-2507:flex