Model specification · Qwen
Qwen3.8 2.4T A95B
qwen/qwen3.8-2.4t-a95b
Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is...
SpecificationsReported values
- Context window
- 1.0M
- Maximum output
- 131K
- Architecture
- text->text
- Tokenizer
- Qwen
- Knowledge cutoff
- —
- Moderated
- No
- Input modalities
- text
- Output modalities
- text
Full pricingUSD where reported
| Price | Standard |
|---|---|
| Input | $2.00 / M |
| Output | $6.00 / M |
Benchmark indexesReported indexes
- Intelligence
- 40
- Coding
- 71.9
- Agentic
- 50.4
- Design Arena Elo
- —
Example workload costsStandard pricing only
| Workload | Input / output | Cache share | Estimated cost |
|---|---|---|---|
| Quick check | 10K / 1K × 1 | 0% | $0.026 |
| Long analysis | 100K / 4K × 1 | 50% | $0.137 |
| Workday | 20K / 2K × 1,000 | 30% | $41.50 |
More from QwenAll Qwen models
| Model | Context | Input / M | Output / M | Intel. |
|---|---|---|---|---|
| Qwen3.8 Max (0902) | 1M | $2.00 | $6.00 | 40.3 |
| Qwen3.8 27B | 1M | $0.420 | $3.00 | 33.9 |
| Qwen3.7 Max | 1M | $1.48 | $4.42 | 29.9 |
| Qwen3.7 Plus | 1M | $0.320 | $1.28 | 25.8 |
| Qwen3.6 27B | 262K | $0.300 | $2.00 | 21.9 |
| Qwen3.5 397B A17B | 262K | $0.550 | $3.50 | 19.1 |
| Qwen3.5-122B-A10B | 262K | $0.260 | $2.08 | 16.2 |
| Qwen3 235B A22B Thinking 2507 | 131K | $0.230 | $2.30 | 12.7 |
| Qwen3 Coder Next | 262K | $0.120 | $0.800 | 10.1 |
| Qwen3 30B A3B Thinking 2507 | 82K | $0.200 | $2.40 | 9.8 |
| Qwen3 32B | 131K | $0.080 | $0.280 | 7.2 |
| Qwen3 14B | 131K | $0.227 | $0.910 | 6.4 |