Source history
Every source file is listed.
Latest snapshot hash, source URL, snapshot date, parser version, check status, and change reason stay visible.
| Source | File hash | Snapshot | Parser | Status | Records | Change reason |
|---|---|---|---|---|---|---|
| Arena | 60b99dc9f6dac066eb0d2adb4ded78c9db0c5c6272f3d0c7920b7b7a27d28138 | 2026-10-03 | 0.6.0 | verified | 23,086 | Parsed 23086 Arena leaderboard records. |
| Artificial Analysis | 7649ce02398ef24f2ec8593949ad0f7bbb55e82c13aed81751348b4be370f946 | 2026-10-03 | 0.5.0 | verified | 6,750 | Parsed 8412 Artificial Analysis public leaderboard records across 868 models and 23 benchmarks. |
| BridgeBench | e88bdde9953c1db7dc3703cb374374f9e117a8ce1421f1f70a195f8cea3766a6 | 2026-05-15 | 0.3.0 | verified | 351 | Loaded 351 live BridgeBench benchmark records. |
| LiveBench | cf57b8354c09ecae3a05e6dd4bab0c95bb432274689c3c753ea7983418ec951e | 2026-10-03 | 0.5.0 | verified | 2,046 | Fetched 66 LiveBench website leaderboard rows for release 2026-06-25 and mapped 2046 benchmark records across 31 benchmark slices. |
| LLMBase | b4e40639ead08ad1025cf15d6014dd72363dcb5532421df0409a050f06ebd187 | 2026-10-03 | 0.2.0 | relay | 0 | Parsed 0 LLMBase intelligence records. |
| MTEB | 23f77e243092b2dc66631c59ae87b05da193d0ed6b92036a4c2401dfd51e1b12 | 2026-04-22 | 0.2.0 | verified | 8 | Loaded 8 verified benchmark records for mteb. |
| OpenCompass | 801864a2daa42afe8a6c87c26fb901fc37dae70c272836655b3f7f118a9af93b | 2026-04-22 | 0.2.0 | verified | 7 | Loaded 7 verified benchmark records for opencompass. |
| Provider official evals | 9604bb00156d076e50019e92b4839563dd1984dca66c28572795fea407f0d812 | 2026-04-19 | unrecorded | verified | 36 | No parser diff has been published for the latest snapshot. |
| Scale Labs | 9210b181118748f99290cf1f03907715e1b00f8989faf84ea3fcd37ccb4164b5 | 2026-10-03 | 0.2.0 | verified | 270 | Loaded 270 live Scale Labs benchmark records. |
| Terminal-Bench | 1eeffc806116555851d93bbc2894350c1f60ede63a4d0126c09e37eb6c998c4f | 2026-10-03 | 0.3.0 | verified | 27 | Loaded 27 Terminal-Bench 4.0 benchmark records from official published rows. |
| Vals AI | 7832d50554658edba8e81fdbcde0aec9caee597a004047e3b66f398373a47ecb | 2026-10-03 | 0.2.0 | verified | 2,039 | Loaded 2039 live Vals AI benchmark records. |