// Ai2 OLMo / Molmo · Text LLM
Olmo-3.1-32B-Instruct: VRAM & local deployment
32.2B parameters. About 23.3 GB at 4-bit and 72.8 GB at BF16 with one 8K-token request.
released 2025-12-10licence: Commercial OKmemory figures are estimates
01Specs
| Publisher | Allen Institute for AI · allenai |
|---|---|
| Family / series | Ai2 OLMo / Molmo · Olmo-3.1 |
| Type | Text LLM (text-generation) |
| Parameters | 32.2B Source: counted from official safetensors weights |
| Context length | 64K tokens (65,536)max_position_embeddings in config.json |
| Architecture | 64 layers (16 full-attention) · 48 sliding-window layers (4096) · GQA |
| Released weights | BF16 · 64.5 GBdownload |
| Licence | Apache-2.0 · Commercial OKPermissive licence (Apache-2.0 / MIT / CC-BY / OpenMDW): commercial use allowed. |
| Released | 2025-12-10 (Hugging Face repo created) |
02How much VRAM it needs (estimate)
Total = weights + 10% + 0.5GB overhead + KV cache for one request (FP16 KV). More concurrent users multiply the KV column.
| Precision | Weights | Total @ 8K | Total @ 32K | Smallest GPU | Mac |
|---|---|---|---|---|---|
| BF16 / FP16 (original) | 64.5 GB | 72.8 GB | 74.4 GB | RTX PRO 6000 96GB | Mac 128GB |
| FP8 / INT8 / Q8_0 | 34.3 GB | 39.5 GB | 41.1 GB | 2 × RTX 5090 (64GB) | Mac 64GB |
| 4-bit (Q4_K_M / AWQ / GPTQ-Int4) | 19.5 GB | 23.3 GB | 24.9 GB | RTX 5090 32GB | Mac 64GB |
KV cache per request: 8K 1.3 GB · 32K 3.0 GB
Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →
03How to run it
Only runtimes the publisher or the runtime’s official library names for this model.
- Transformers — mentioned in the official model card
- Ollama — in the official Ollama library · olmo-3.1:32b
ollama run olmo-3.1:32b04Official repositories & quantised builds
1 official repo(s) from allenai.
- Olmo-3.1-32B-Instruct Instruct
05Other Ai2 OLMo / Molmo models
All 31 →| Model | Params | 4-bit total | Released |
|---|---|---|---|
| Molmo2-ER | 4.85B | 4.9 GB | 2026-05-04 |
| MolmoWeb-4B | 4.85B | 4.9 GB | 2026-03-20 |
| MolmoPoint-8B | 8.68B | 7.5 GB | 2026-03-16 |
| Olmo-Hybrid-7B | 7.43B | 6.5 GB | 2026-01-28 |
| Molmo2-4B | 4.85B | 4.9 GB | 2025-12-14 |
| Molmo2-8B | 8.66B | 7.5 GB | 2025-12-14 |
| Molmo2-O-7B | 7.76B | 10.0 GB | 2025-12-14 |
| Olmo-3.1-32B-Think | 32.2B | 23.3 GB | 2025-12-10 |
06Sources
Data checked 2026-01-05. Parameters, context and licence are copied from the official repository; memory figures are our estimates.