Sign in

Qwen3.5-35B-A3B

Qwen

Hugging FaceArtificial Analysis

BenchClub cloud

Pass rate
96%
26 of 27 checks
Speed
201.4 tok/s
decode, short prompt
At 8k context
173.2 tok/s
decode
Cloud runs
3
speeds within 2.5%
  • Coding9/9100%
  • Tool use8/989%
  • Long context9/9100%
Suite
[email protected]
Machine
NVIDIA L40S · 48 GB
Engine
ollama 0.40.14

Headline numbers come only from cloud runs. Runs on members’ own machines are listed below and don’t change them.

Arena score
1395
Among open models
#46of 144
Among all on Arena
#165
Votes
30,427

Measured on Qwen3.5-35B-A3B, the served version, not a local build. LMArena, CC BY 4.0, published Oct 28, 2026.

Score by category

CategoryScore
  • Chinese1456
  • Medicine & health1421
  • Software & IT1415
  • French1413
  • Science1413
  • Hard prompts, English1410
  • Applied math1410
  • Coding1410
  • Expert1407
  • English1407
  • Math1404
  • Hard prompts1400
  • Legal & government1398
  • Business & finance1393
  • Multi-turn1390
  • Spanish1390
  • Long prompts1389
  • Polish1385
  • Instructions1379
  • Non-English1378
  • Russian1377
  • Excluding ties1373
  • German1369
  • Writing & literature1365
  • Korean1353
  • Creative writing1347
  • Media & sports1347
  • Japanese1325

BenchClub runs

Run this model
Run by

PassedFailedNot supported