by eleutherai
Llemma 7B is a language model for mathematics. It was initialized with Code Llama 7B weights, and trained on the Proof-Pile-2 for 200B tokens. Llemma models are particularly strong at chain-of-thought mathematical reasoning and using computational tools for mathematics, such as Python and formal theorem provers.
| Signal | Strength | Weight | Impact |
|---|---|---|---|
| Recency2026-03-03T20:27:32.487Z | 74 | 15% | +11.1 |
| Context Window2026-03-03T20:27:32.487Z | 57 | 15% | +8.6 |
| Output Capacity2026-03-03T20:27:32.487Z | 60 | 10% | +6.0 |
| Capabilities2026-03-03T20:27:32.487Z | 14 | 25% | +3.6 |
| Versatility2026-03-03T20:27:32.487Z | 33 | 10% | +3.3 |
| Pricing Tier2026-03-03T20:27:32.487Z | 1 | 25% | +0.3 |
Cost Estimator
You save $31.59/month vs category average