MMLU (HELM)
Mirrored from HELM. Licence: Apache-2.0. HELM (Stanford Center for Research on Foundation Models), stanford-crfm/helm.
Run it yourself
On openvibe.codes, with your own keys or the free pool through OpenVibe.AI. The result is published here.
- Tasks
- 1
- Judged by
- —
- Limits
- —
- Verified runs
- 0
Rankings
No verified runs yet
A run counts here once openvibe.codes, Run or a Node worker ran it and published it. Submitted runs are listed below but never ranked.
Tasks
1- 1mmlu
Mirrored result: MMLU.
Latest runs
Compare the newest two- Palmyra-X-004HELM v1.13.0 · 73.9/100SubmittedMirrored
- Palmyra X V3 (72B)HELM v1.13.0 · 70.2/100SubmittedMirrored
- Palmyra X V2 (33B)HELM v1.13.0 · 62.1/100SubmittedMirrored
- Solar ProHELM v1.13.0 · 67.9/100SubmittedMirrored
- GPT-4o mini (2024-07-18)HELM v1.13.0 · 66.8/100SubmittedMirrored
- GPT-4o (2024-08-06)HELM v1.13.0 · 73.8/100SubmittedMirrored
- GPT-4o (2024-05-13)HELM v1.13.0 · 74.8/100SubmittedMirrored
- GPT-4 Turbo (2024-04-09)HELM v1.13.0 · 71.1/100SubmittedMirrored
Badges and embeds
A badge for your README, and a watch that tells you when this benchmark moves.
Sign in to follow this, and Bench will tell you when it moves.
The same data as JSON: definition · rankings