Olmo-3.1-32B-Think
by allenai
32.2B params · text-generation · 86 likes · 10.4k downloads
Olmo-3.1-32B-Think is a 32.2B parameter model. At Q4 quantization it requires 16GB of VRAM. It runs comfortably on GeForce RTX 4090 (40 tok/s), GeForce RTX 5090 (61 tok/s), A100 PCIe 80 GB (74 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| Public AI | 66 tok/s |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| GeForce RTX 4090 | 24GB | 40 tok/s | comfortable |
| GeForce RTX 5090 | 32GB | 61 tok/s | comfortable |
| M4 Max 128GB | 128GB | 14 tok/s | tight |
| M4 Pro 48GB | 48GB | 7 tok/s | tight |
| M4 Pro 24GB | 24GB | 7 tok/s | tight |
| A100 PCIe 80 GB | 80GB | 74 tok/s | comfortable |
| H100 SXM5 80 GB | 80GB | 150 tok/s | comfortable |
| GeForce RTX 3090 | 24GB | 36 tok/s | comfortable |
| Radeon RX 7900 XTX | 24GB | 29 tok/s | tight |
| M4 Max 64GB | 64GB | 14 tok/s | tight |