Alibaba

Qwen3.8 2.4T A95B

Aug 12, 2026ReasoningOpen weightsModels.devHugging Face

Compare

Intelligence

39.9

#64 of 744Rank 64 of 744 measured models

Coding

71.9

#43 of 298Rank 43 of 298 measured models

Output speed

39.5 t/s

#366 of 403Rank 366 of 403 measured models

Price/1M

$3

#487 of 659Rank 487 of 659 measured models

Context window

262K

#458 of 1199Rank 458 of 1199 measured models

Cost per task

$2.16

#166 of 193Rank 166 of 193 measured models

Qwen3.8 2.4T A95B scores 39.9 on the Artificial Analysis Intelligence Index, ranking 64th of 744 measured models. The closest models on this index are Qwen3.8 Max (40.2), Qwen3.8-Flash-Next (39.8), and Muse Spark 1.2 (Xhigh) (39.6). At $3 per million tokens (3:1 input/output blended price), it is priced close to Qwen3.8 Max ($3), more expensive than Qwen3.8-Flash-Next ($0.23) and Muse Spark 1.2 (Xhigh) ($2). With an output speed of 39.5 t/s, it is about as fast as Qwen3.8 Max (40.1 t/s), slower than Qwen3.8-Flash-Next (51.8 t/s) and Muse Spark 1.2 (Xhigh) (322 t/s).