#10Alibaba· released 2026-08-12

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...

View on OpenRouter ↗
Context
1.0M
Input $/1M
$2.00
Output $/1M
$6.00
Modality
text
Arena Elo
1588
LMArena — human blind votes
AA Index
64.0
Artificial Analysis composite
LiveBench
86.7
Contamination-resistant
GPQA
88.4%
Grad-level science reasoning
MMLU-Pro
88.7%
Broad knowledge
SWE-bench
74.8%
Real GitHub bugs (Verified)
Terminal-Bench
65.2%
Agentic shell tasks
ARC-AGI
32.8%
Abstract reasoning
Aider Polyglot
82.4%
Multi-language code editing
Speed
77 tok/s
Median output speed

Profile

Qwen: Qwen3.8 2.4T A95B is Alibaba's entry at position #10 on the current aggregated leaderboard. Its strongest signals are graduate-level science reasoning, real-world software engineering, unattended agentic work, and very long context. Every figure below is aggregated from the public sources listed on the homepage and refreshed on each pipeline run.

The score spread is comparatively complete for this model, but the standing advice still applies: treat gaps of a few points as ties and validate on your own workload before committing.

How this model's position is computed is documented in the methodology; what each benchmark actually measures is in the benchmark guide. See a figure that disagrees with its source? Report it — accepted corrections apply on the next refresh.

Nearby on the leaderboard