Qwen3 32B
Dense open Qwen model for self-hosted chat, reasoning, and coding
Cheapest input
$0.08
$/MTok · models.dev
Cheapest output
$0.28
$/MTok
Context
131K
tokens
Offers
15
3 sources
Price changes
Cheapest list input/output only. Not a daily candle chart: rates move rarely, so this is an event log. Arrow down means cheaper.
- ↑ riseOutput $0.24 → $0.28
Offers
Same model, different listings. First-party, router, and aggregator rows stay separate so you can see the spread instead of a single blended number.
| Source | Provider | Input | Output | Cache read | Context |
|---|---|---|---|---|---|
| models.devcheapestQwen/Qwen3-32B | DeepInfra | $0.08 | $0.28 | $0 | 41K |
| OpenRouterqwen/qwen3-32b | OpenRouter | $0.08 | $0.28 | $0 | 131K |
| models.devQwen/Qwen3-32B | Abacus | $0.09 | $0.29 | $0 | 131K |
| models.devqwen/qwen3-32b | NanoGPT | $0.1 | $0.3 | $0 | 41K |
| models.devQwen/Qwen3-32B | Nebius | $0.1 | $0.3 | $0.01 | 128K |
| LiteLLMQwen/Qwen3-32B | — | $0.1 | $0.28 | $0.01 | — |
| LiteLLMdeepinfra/Qwen/Qwen3-32B | — | $0.1 | $0.28 | $0.01 | — |
| models.devqwen/qwen3-32b | Kilo Gateway | $0.104 | $0.416 | $0 | 41K |
| models.devQwen/Qwen3-32B | SiliconFlow (China) | $0.14 | $0.57 | $0 | 131K |
| models.devQwen/Qwen3-32B | SiliconFlow | $0.14 | $0.57 | $0 | 131K |
| models.devqwen/qwen3-32b | Merge Gateway | $0.15 | $0.6 | $0 | 131K |
| models.devqwen/qwen3-32b | Groq | $0.29 | $0.59 | $0.029 | 131K |
| models.devQwen/Qwen3-32B | Hugging Face | $0.29 | $0.59 | $0 | 131K |
| LiteLLMqwen/qwen3-32b | — | $0.29 | $0.59 | $0.029 | — |
| models.devQwen/Qwen3-32B | Pioneer | $0.9 | $0.9 | $0.9 | 131K |
Capabilities
- Reasoning
- Tool calling
- Structured output
- Attachments
- Open weights
- In: text
- In: pdf
- Out: text
Knowledge cutoff: 2025-04Released: 2025-04Family: qwen
Scores
Sparse coverage on purpose. Missing a score means we have not linked one yet, not that the model is unranked. Full boards on /benchmarks.
- AA intelligence11.5
- AA coding15.3
- AA agentic1.8
Sources: Arena AI, Artificial Analysis. See /benchmarks.