Qwen3 32B Thinking Now Live on TheRouter — Dense Reasoning at $0.56/M Input
Qwen3 32B Thinking is the dense-architecture reasoning variant from the Qwen3 family. 32 billion parameters, explicit thinking mode, 128K context, served via qwen (CN-region partner capacity) at $0.56/M input and $5.60/M output. A solid pick for reasoning-heavy workloads that don't need the price/quality jump to 235B MoE.
Why dense over MoE here
- Predictable latency — no expert-routing variance per token, useful for SLO-bound services.
- Consistent memory footprint upstream — fewer cold-start tail effects on the partner side.
- Reasoning-tuned with explicit chain-of-thought surfaced in
completion_tokens_details.reasoning_tokens. - 128K context at flat pricing — no tiered surprises.
Pricing & capabilities
| Property | Value |
|---|---|
| Standard alias | qwen/qwen3-32b-thinking |
| Input price | $0.56 / 1M tokens |
| Output price | $5.60 / 1M tokens |
| Context | 128,000 tokens |
| Capabilities | Streaming · Reasoning · JSON mode |
| Served via | qwen (CN-region partner capacity) |
Prices shown are launch-time figures; see each model page for the current billed rate.
Try it
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen/qwen3-32b-thinking",
"messages": [{"role": "user", "content": "Solve: a chicken and a half lays an egg and a half in a day and a half. How many eggs do 9 chickens lay in 9 days?"}],
"provider": {"only": ["qwen"]}
}'