1 source, 1 independent of the vendor: centre 12; fewer than 3 sources or none independent, so shown as a band.
- Artificial Analysis · third party
- charxiv rq: 81.3 · vendor · not used: only 0 reference models
- claweval mm: 59.9 · vendor · not used: no reference models for this benchmark
- mathvision: 84.87 · vendor · not used: no reference models for this benchmark
- mmmu pro: 79 · vendor · not used: no reference runs on the same harness
- mmsearch: 79 · vendor · not used: no reference models for this benchmark
- omnidocbench 1 5: 91.35 · vendor · not used: no reference models for this benchmark
- webvoyager: 90.83 · vendor · not used: no reference models for this benchmark
how FTI is computed