OLMo · Allen Institute for AI
Fine-tuning OLMo 3 32B
What does it take to fine-tune OLMo 3 32B?
OLMo 3 32B needs 64 GB for half-precision LoRA training and 48 GB in four-bit, and 64 GB to serve. It supports supervised fine-tuning, under the Apache 2.0 licence. The cheapest qualifying class is A40 at $0.42 per GPU-hour.
The strongest fully-open model at 32B scale. Same memory profile as Qwen3 32B, so the choice between them is not about cost — it is about whether published training provenance is worth more to you than the surrounding ecosystem.
What to know before choosing it
The strongest fully-open model at 32B scale. Its memory profile is identical to Qwen3 32B, so the choice between them is not about cost — it is about whether published training provenance is worth more to you than the larger surrounding ecosystem.
For most commercial work Qwen3 will be the easier path. For work that has to be auditable end to end, this is one of very few options at this scale, and the trade is worth making deliberately.
Specification
| Parameters | 32B |
|---|---|
| Architecture | Dense |
| Size tier | Large |
| LoRA training memory | 64 GBHalf precision, frozen base |
| QLoRA training memory | 48 GBFour-bit base, higher-precision adapter |
| Serving memory | 64 GBHalf precision, before attention cache |
| Objectives | SFT |
| Licence | Apache 2.0 |
| Repository | allenai/OLMo-3-32B |
What it costs to train
Every class with enough memory for four-bit training, cheapest first. Total cost is the rate multiplied by wall time, so the cheapest rate is not always the cheapest run — a faster class that finishes sooner frequently wins.
| GPU class | VRAM | Training / hr | Serving / hr | Fits |
|---|---|---|---|---|
| A40 | 48 GB | $0.42 | $0.48 | QLoRA only |
| RTX 6000 Ada | 48 GB | $0.61 | $0.71 | QLoRA only |
| A6000 | 48 GB | $0.65 | $0.75 | QLoRA only |
| L40S | 48 GB | $0.78 | $0.90 | QLoRA only |
| A100 80 GB | 80 GB | $1.68 | $1.94 | LoRA and QLoRA |
| H100 80 GB | 80 GB | $2.59 | $2.99 | LoRA and QLoRA |
| H200 | 141 GB | $4.55 | $5.25 | LoRA and QLoRA |
Half-precision LoRA needs 64 GB, so the cheapest class for it is A100 80 GB at $1.68 per hour. Below that, training has to be quantised.
Serving fits on A100 80 GB at $1.94 per hour — before the attention cache, which grows with context length and concurrency.
Good starting point for
- Large-scale work with auditable provenance
- Public-sector and research deployments
Other sizes in this family
Comparable sizes elsewhere
Frequently asked questions
How much VRAM does it take to fine-tune OLMo 3 32B?
64 GB for half-precision LoRA and 48 GB for four-bit QLoRA. Serving needs 64 GB. Preference tuning roughly doubles the training figure, because a frozen reference model is held alongside the one being trained.
What is the cheapest way to fine-tune OLMo 3 32B?
Four-bit QLoRA on A40 at $0.42 per GPU-hour is the cheapest class that meets the 48 GB threshold. Note that quantised training is slower per step, so a faster class sometimes costs less over the whole run.
Can I download the weights after fine-tuning OLMo 3 32B?
Yes. Every finished run exposes its trained weights for download, and publishing to a model hub is a single call with a generated model card recording the base model and version the adapter applies to.
What licence does OLMo 3 32B carry?
Apache 2.0. The licence follows the fine-tune — a derivative inherits the base model’s terms, and those terms pass to anyone you give the model to.
Last verified 6 August 2026. Memory thresholds are the platform's own admission limits.
Fine-tune OLMo 3 32B
Upload a dataset, forecast the run, and see the cost before any compute is leased.