modelbenchmark.io

Qwen3 235B A22B Thinking 2507

Alibaba · qwen3-235b-a22b-thinking

Compare

Reasoning model for deliberate analysis, multi-step problem solving, and tool use

Specification

most-agreed values

Context
262K
Max output
4K
Released
2025-08-12
Knowledge cutoff
2025-04
Retires
2026-04-16
Open weights
no
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.23
Output
$2.30
Cache read
$0.055
Cache write
$0.22

Available from 22 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
iflowcnfreefree256K64K
modelscopefreefree262K131K
wandb · wandb$0.10$0.10262K262K
io-net$0.11$0.60$0.055$0.22262K4K
siliconflow$0.13$0.60262K262K
siliconflow-cn$0.13$0.60262K262K
submodel$0.20$0.60262K131K
fireworks_ai$0.22$0.88262K262K
kilo$0.23$2.30131K118K
openrouter$0.23$2.30
131K
262K
118K
262K
deepinfra$0.30$2.90262K262K
helicone$0.30$2.90262K82K
huggingface$0.30$3.00262K131K
jiekou$0.30$3.00131K131K
llmgateway$0.30$3.00262K8K
llmgateway-providers$0.30$3.00131K33K
novita$0.30$3.00131K33K
novita-ai$0.30$3.00131K33K
vercel$0.40$4.00131K33K
venice$0.45$3.50128K16K
together_ai$0.65$3.00256K2026-04-16
qiniu-ai262K4K

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02openrouterInput$0.23
2026-09-06openrouterInput$0.23
2026-09-02openrouterOutput$2.30
2026-09-06openrouterOutput$2.30

More from Alibaba

most-hosted first