QwenOpen weightsReleased 12 Aug 2026
qwen/qwen3.8-2.4t-a95bPeer group: 34 models with a blended list price over $1 per 1M tokens (3 input : 1 output). Charts below compare Qwen3.8 2.4T A95B with all models; cost and speed charts focus on this peer group.
Verification score: the share of 80 questions from open datasets answered correctly. See the benchmark datasets.
Qwen3.8 2.4T A95B circled; dashed line joins models no cheaper model beats
Qwen3.8 2.4T A95B circled · accurate and quick sits top left
Qwen3.8 2.4T A95B circled · output tokens, reasoning included
All models tested on the current suite
Average cost of one question, split into input, reasoning and answer tokens at each model's list price. Reasoning tokens are billed as output.
Peer group: blended price over $1
All models · lower is better
Median of 3 streamed runs of the same prompt, measured 25 Oct 2026.
Peer group (over $1) · higher is better
Peer group (over $1) · lower is better
Peer group (over $1) · lower is better
Qwen3.8 2.4T A95B circled · quick to start and fast sits top left
Questions answered correctly in each category
Head to head with its closest rivals, other Qwen models and the top models overall.
Every host serving this model through OpenRouter, from OpenRouter's catalogue
| Host | Input / 1M | Output / 1M | Context | Max output | Quantization |
|---|---|---|---|---|---|
| Novita | $2 | $6 | 1M | 131k | — |
| Alibaba | $2 | $6 | 1M | 131k | — |
| SiliconFlow | $2 | $6 | 1M | 131k | fp8 |
| Venice | $2 | $6 | 262k | 66k | — |
| Together | $2 | $6 | 1M | 909k | — |
Scores, speed and billed cost are our own measurements. Prices, context window and hosts come from OpenRouter's catalogue on the test date.