All models
Alibaba (Qwen)
Qwen3 32B fastruns locally
alibaba/qwen3-32b
Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for...
Input / 1M
$0.080
USD per million tokens
Output / 1M
$0.28
USD per million tokens
Est. cost / token
$0.00000013
$0.13 per 1M, 3:1 mix
Context window
41K
max output 16K
Major features
Open weightsReasoningJSON schemaTools
Task suitability (heuristic 1–5, not a benchmark)
- coding3
- reasoning3
- math3
- summarization4
- extraction4
- classification4
- chat3
- creative3
- translation4
- vision1
- long context2
- agentic3
Run it locally
Will it run on my machine? →ollama run qwen3:32b| Quantization | Quality | Memory needed (8K ctx) |
|---|---|---|
| Q2_K | poor | 17 GB |
| Q3_K_M | fair | 19.4 GB |
| Q4_K_Mrecommended | good | 23.6 GB |
| Q5_K_M | very good | 27 GB |
| Q6_K | excellent | 30.8 GB |
| Q8_0 | near-lossless | 39.1 GB |
| F16 | lossless | 71.4 GB |
Estimates.
Price history
| Recorded | Input | Output | Cached | Source |
|---|---|---|---|---|
| 2026-11-05 03:20 | $0.080 | $0.28 | — | openrouter |
A row is added whenever a sync detects a price change. Vendor pricing source
Pricing (USD per 1M tokens)
- Input
- $0.080
- Output
- $0.28
- Blended (3:1)
- $0.13
Details
- Company
- Alibaba (Qwen)
- Version
- 3
- Family
- Qwen
- Released
- 2025-04-28
- Parameters
- 32.8B
- Max output
- 16K
- Open weights
- Yes
- Inputs
- text
- Last synced
- 25 min ago