priorsBuilder

Alibaba · released 2026-08-13

Qwen 3.8 2.4T-A95B

Measured on 28 benchmarks.

General frontier

44.8

#44 of 136 models

includes estimates: 38% of the mix is measured for this model

Your mix’s benchmarks

3 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #64default of 58742.4%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #109default of 5264.3

Leader: Opus 5.5 · 46.4

reasoning

SimpleBench

97 results
  1. #28default of 9762.5%

Leader: Opus 5.5 · 88.4%

All 28 benchmarks

show 25 more

agentic

AA-Briefcase v1.1

224 results
  1. #50default of 2241435

Leader: Sonnet 5.5 · 1823

agentic

AutomationBench-AA

219 results
  1. #46default of 21957.2%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #8default of 5047.4%

Leader: Fable 5 · 51.1%

agentic

τ³-Banking

206 results
  1. #5default of 20649.1%

Leader: Qwen 3.8 Max · 51.3%

coding

SciCode

226 results
  1. #68default of 22654.1%

Leader: Opus 5.5 · 66.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #39default of 23882.0%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #84default of 21611.1%

Leader: Sonnet 5.5 · 63.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #60default of 61039.9

Leader: Opus 5.5 · 57.6

cost

AA Intelligence Index Cost per Task

177 results
  1. #151default of 177$2.16

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #148default of 177$2,633

Leader: GPT-6 Luna · $10.63

economics

GDP.pdf

210 results
  1. #95default of 21015.0%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #28default of 2881610

Leader: Opus 5.5 · 1866

knowledge

AA-Omniscience Accuracy

526 results
  1. #169default of 52631.3%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #69default of 52660.8%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

GPQA Diamond

539 results
  1. #18default of 53993.5%

Leader: GPT-6 Astra · 96.3%

long-context

AA-LCR v1.1

523 results
  1. #71default of 52380.3%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #92default of 940.0%

Leader: Sonnet 5.5 · 75.0%

reasoning

CritPt

529 results
  1. #60default of 52920.0%

Leader: Muse Wonder Full (Max) · 32.9%

reliability

Tool-call error rate

171 results
  1. #71default of 1711.0%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #92default of 18199.5%

Leader: GPT-5 Pro · 100.0%

safety

CWE-Bench

19 results
  1. #17default of 1928.3%

Leader: Grok 4.7 · 68.3%

speed

Latency (Time to First Token)

351 results
  1. #206default of 3513.1s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #326default of 35140.0

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #97default of 1771.4s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #79default of 17670.8

Leader: Muse Spark 1.1 · 204.3