← Model list

Qwen3-Next 80B-A3B (Thinking)

alibaba·alibaba/qwen3-next-80b-a3b-thinking·GA·Open weights·qwen series

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

At a glance
Official price$0.144 / $1.43 per 1M
Cache read · Alibaba (China)
Lowest paid$0.15 / $0.65
NanoGPT Gateway · 3.5× spread
Context131,072
⚠ Providers report 65,536–262,144; the table below is authoritative
Output limit32,768
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
Text
Knowledge cutoff2025-04
Released / updated2025-09 / 2025-09

Quality & performance

Artificial Analysis doesn't cover this model (267 of 2059 have data). Quality data comes from independent evals covering widely used models.

Available at 16 providers15 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
NanoGPTGateway$0.15$0.65$0.075256,00032,768
CortecsGateway$0.149$1.20128,000128,000
OpenRouterGateway$0.15$1.20262,14432,768
Nebius Token Factory
Qwen/Qwen3-Next-80B-A3B-Thinking
Cloud$0.15$1.20$0.015$0.18128,00016,384
Merge GatewayGateway$0.15$1.20131,07232,768
Vercel AI GatewayCloud$0.15$1.20131,07232,768
Eden AIGateway$0.15$1.20131,07232,768
DevPass (LLM Gateway)Gateway$0.15$1.20131,07232,768
Kilo GatewayGateway$0.15$1.20131,07232,768
Alibaba (China)OfficialFirst-party$0.144$1.43131,07232,768

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1NanoGPT$53.50
2Nebius Token Factory$73.80
3Cortecs$89.55
4OpenRouter$90.00
5Merge Gateway$90.00
6Vercel AI Gateway$90.00
10Alibaba (China) · Official$100.50
Switch to NanoGPT to save $47.00/mo (47%)
Note: this is a gateway; verify availability and rate limits yourself.

Benchmark

No upstream benchmark data for this model. For quality, see the Artificial Analysis intelligence score above.

Reasoning control

budget_tokenseffort = minimaleffort = loweffort = mediumeffort = higheffort = xhigheffort = maxToggle (on / off)

8 / 16 providers expose no reasoning control (reasoning_options: []).

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$6.00$02026-08-052026-08-132026-08-05 · Input list $0.502026-08-06 · Input list $0.502026-08-07 · Input list $0.502026-08-08 · Input list $0.502026-08-09 · Input list $0.502026-08-10 · Input list $0.502026-08-11 · Input list $0.502026-08-12 · Input list $0.502026-08-13 · Input list $0.1442026-08-05 · Output list $6.002026-08-06 · Output list $6.002026-08-07 · Output list $6.002026-08-08 · Output list $6.002026-08-09 · Output list $6.002026-08-10 · Output list $6.002026-08-11 · Output list $6.002026-08-12 · Output list $6.002026-08-13 · Output list $1.432026-08-05 · Min blended $0.2752026-08-06 · Min blended $0.2752026-08-07 · Min blended $0.2752026-08-08 · Min blended $0.2752026-08-09 · Min blended $0.2752026-08-10 · Min blended $0.2752026-08-11 · Min blended $0.2752026-08-12 · Min blended $0.2752026-08-13 · Min blended $0.275