Qwen3-Max-Thinking
by ZenMux · listed Jan 2026RT
Qwen reasoning model for deliberate problem solving, math, and coding
Current rates USD per 1M tokens
Input$1.20
Output$6.00
Cache read—
Cache write—
Against the market
On the record
- Model id
qwen/qwen3-max- Context window
- 256K tokens
- Max output
- 64K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- no
- Knowledge cutoff
- Jan 2025
- Released
- 2026-01-23
- Last updated
- 2026-01-23
- Provider docs
- docs.zenmux.ai
Back of the envelope
≈
Also listed at
- iFlowfree in · free out
- OrcaRouter$0.359 in · $1.43 out
- OpenRouter$0.78 in · $3.90 out
- LLM Gateway$0.845 in · $3.38 out
- Alibaba (China)$0.861 in · $3.44 out
- NanoGPT$1.08 in · $5.40 out
- Vercel AI Gateway$1.20 in · $6.00 out
- Abacus$1.20 in · $6.00 out
- Deep Infra$1.20 in · $6.00 out
- Kilo Gateway$1.20 in · $6.00 out
- Alibaba$1.20 in · $6.00 out
- NovitaAI$2.11 in · $8.45 out
- Qiniu— in · — out