Qwen3-8B
Qwen
BenchClub cloud
- Pass rate
- 89%
- 32 of 36 checks
- Speed
- 120.7 tok/s
- decode, short prompt
- At 8k context
- 94.2 tok/s
- decode
- Cloud runs
- 4
- speeds within 2.7%
- Coding9/1275%
- Tool use11/1292%
- Long context12/12100%
- 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.
LMArena
Not ranked on LMArena
LMArena only ranks the models it serves.
BenchClub runs
Run this modelRun byWhereSpeed
Ben CarterNov 6, 2026RTX 4090 · 24 GB141.5 tok/s
Diego HernándezNov 5, 2026RTX 3090 · 24 GB124.3 tok/s
Alex RiveraNov 3, 2026L40S · 48 GBCloud121.7 tok/s
Saoirse ByrneNov 2, 2026Apple M4 Max · 128 GB65.7 tok/s
- Li WeiOct 29, 2026L40S · 48 GBCloud119.1 tok/s
- Niko LaineOct 17, 2026L40S · 48 GBCloud119.7 tok/s
- Kwame MensahOct 11, 2026Apple M2 Pro · 32 GB25.1 tok/s
- Niko LaineOct 9, 2026RTX 5090 · 32 GB251.0 tok/s
Samir HaddadOct 1, 2026Apple M4 Max · 128 GB68.4 tok/s
PassedFailedNot supported