Model specification · Qwen
Qwen3 8B
qwen/qwen3-8b
Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue. It supports seamless switching between "thinking" mode for math,...
SpecificationsReported values
- Context window
- 131K
- Maximum output
- 8.2K
- Architecture
- text->text
- Tokenizer
- Qwen3
- Knowledge cutoff
- 2025-03-31
- Moderated
- No
- Input modalities
- text
- Output modalities
- text
Full pricingUSD where reported
| Price | Standard |
|---|---|
| Input | $0.117 / M |
| Output | $0.455 / M |
Benchmark indexesReported indexes
- Intelligence
- 8.3
- Coding
- 9
- Agentic
- 1.5
- Design Arena Elo
- —
Example workload costsStandard pricing only
| Workload | Input / output | Cache share | Estimated cost |
|---|---|---|---|
| Quick check | 10K / 1K × 1 | 0% | $0.00162 |
| Long analysis | 100K / 4K × 1 | 50% | $0.014 |
| Workday | 20K / 2K × 1,000 | 30% | $3.25 |
More from QwenAll Qwen models
| Model | Context | Input / M | Output / M | Intel. |
|---|---|---|---|---|
| Qwen3.7 Max | 1M | $1.48 | $4.42 | 46 |
| Qwen3.6 Plus | 1M | $0.325 | $1.95 | 39.6 |
| Qwen3.7 Plus | 1M | $0.320 | $1.28 | 39 |
| Qwen3.6 27B | 262K | $0.289 | $2.40 | 37.1 |
| Qwen3.5 397B A17B | 262K | $0.390 | $2.34 | 33.7 |
| Qwen3.5-122B-A10B | 262K | $0.260 | $2.08 | 32.3 |
| Qwen3.6 35B A3B | 262K | $0.140 | $1.00 | 31.6 |
| Qwen3.5-35B-A3B | 262K | $0.140 | $1.00 | 24 |
| Qwen3.5-9B | 262K | $0.100 | $0.150 | 21.4 |
| Qwen3 Coder Next | 262K | $0.120 | $0.800 | 21.1 |
| Qwen3 235B A22B Thinking 2507 | 262K | $0.230 | $2.30 | 19.6 |
| Qwen3 Next 80B A3B Thinking | 262K | $0.150 | $1.20 | 16.7 |