wizardlm-70b
1,194±20
1,174CI1,213
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,318
- Last sync
- 30 min30 minutes ago
All sources syncing normally
- Arena · Text· 7 hours ago
- Arena · WebDev· 7 hours ago
- Arena · Vision· 7 hours ago
- Arena · Agent· 7 hours ago
- GPQA Diamond· 30 minutes ago
- SWE-bench· 30 minutes ago
- Terminal-Bench· 30 minutes ago
- ARC-AGI-2· 30 minutes ago
Filter by org:
Released:
One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.
| Model | Org | ± CI | Swing | Add to compare | |||||
|---|---|---|---|---|---|---|---|---|---|
| 348 | wizardlm-70b | microsoft | Aug 2023 | — | 1,194 | 1,174–1,213 | #350–718 |
Showing Coding exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.