Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
CohereLabs
|
|---|---|
| Company | CohereLabs |
| Release Date | 2026-05-20 |
| Parameters (B) | 218 |
| Architecture | Sparse MoE Transformer |
| Context Window | 128000 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 13 |
| VRAM (GB) | 320 |
| Compute | 4× NVIDIA H100 80GB (Hopper) or 2× NVIDIA B200 (Blackwell); vLLM with tensor parallelism recommended |
| Benchmark: AIME-2025 | 90.4 |
| Benchmark: GPQA-Diamond | 74.4 |
| Benchmark: GSM8K | 95.0 |
| Benchmark: HLE | 11.4 |
| Benchmark: IFEval | 84.3 |
| Benchmark: MATH-500 | 89.8 |
| Benchmark: MMLU | 82.8 |
| Benchmark: MMLU-Pro | 77.8 |
| Benchmark: MMMU | 75.1 |
| Benchmark: MMMU-Pro | 63 |
| Benchmark: Terminal-Bench-2.1 | 22.8 |
| Pricing (Input) | — |
| Pricing (Output) | — |