Qwen3.8 2.4T A95B
NEWQwen · Open weights · Released Aug 12, 2026 · qwen/qwen3.8-2.4t-a95b
1.05M contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 6 providersUSD / MTOK · VERIFIED 4 MINUTES AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
 DeepInfra | $2.00 | $6.00 | $0.20 | 262K | fp4 | 1.3s | 76 tok/s | 4/5 | 96.9% |
 SiliconFlow | $2.00 | $6.00 | $0.25 | 1.05M | fp8 | 2.7s | 49 tok/s | 5/5 | 100.0% |
| MOModal | $2.00 | $6.00 | $0.25 | 1M | nvfp4 | 1s | 122 tok/s | 4/5 | 99.8% |
 Alibaba | $2.00 | $6.00 | $0.25 | 1M | — | 2s | 60 tok/s | 5/5 | 100.0% |
 Together | $2.50 | $6.25 | $0.50 | 1.01M | — | 1.2s | 191 tok/s | 4/5 | 99.7% |
 Venice | $2.50 | $7.50 | $0.313 | 262K | — | 1.5s | 168 tok/s | 3/5 | 100.0% |
TTFT/TPS: median (P50) over trailing 72h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 43 hours ago
Pricing variants · USD / MTok in / out
STANDARD
$2.00 / $6.00
per MTok in / out
CACHE READ $0.20
About
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, with 95 billion active parameters out of 2.4 trillion total. It is...
Specifications
Context window1,048,576
Max output131,072
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedAug 12, 2026
TokenizerQwen
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/qwen/qwen3.8-2.4t-a95b)Sources
OpenRouter API4 minutes ago
AGGREGATORHugging Face4 minutes ago
COMMUNITYModelIndex benchmarks43 hours ago
MEASUREDVerified 4 minutes ago · list prices, not negotiated rates.