modelbenchmark.io

Qwen3-Next 80B-A3B (Thinking)

Alibaba · qwen3-next-80b-a3b-thinking

Compare

Efficient Qwen thinking model for local reasoning, math, and coding agents

Specification

alibaba listing

Context
131K
Max output
33K
Released
2025-09
Knowledge cutoff
2025-04
Retires
2026-02-25
Open weights
no
Input
text
Output
text

Price

US dollars per million tokens · alibaba list price

Input
$0.50
Output
$6.00
Cache read
$0.075

Available from 23 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
alibabavendor$0.50$6.00131K33K
deepinfra$0.14$1.40262K262K
alibaba-cn$0.144$1.434131K33K
dashscope · dashscope$0.15$1.20262K66K
edenai$0.15$1.20131K33K
jalapeno$0.15$1.50131K33K
jiekou$0.15$1.5066K66K
kilo$0.15$1.20131K33K
llmgateway$0.15$1.20131K33K
llmgateway-providers · vertex-openai$0.15$1.20131K33K
merge-gateway$0.15$1.20131K33K
nano-gpt$0.15$0.65$0.075256K33K
nebius$0.15$1.20128K128K
novita$0.15$1.50131K33K
novita-ai$0.15$1.50131K33K
openrouter$0.15$1.20262K33K
qwen_ai_platform · qwen-ai-platform$0.15$1.20262K66K
qwencloud · qwencloud$0.15$1.20262K66K
together_ai$0.15$1.50262K2026-02-25
vercel$0.15$1.20131K33K
huggingface$0.30$2.00262K131K
fireworks_ai$0.90$0.904K4K
qiniu-ai131K33K

Price history

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

DateHostField$/M
2026-09-02nebiusInput$0.15
2026-09-06nebiusInput$0.15
2026-09-02openrouterInput$0.15
2026-09-06openrouterInput$0.15
2026-09-02nebiusOutput$1.20
2026-09-06nebiusOutput$1.20
2026-09-02openrouterOutput$1.20
2026-09-06openrouterOutput$1.20

More from Alibaba

most-hosted first