# Llama 3.1 Nemotron 70B Instruct HF

Llama 3.1 Nemotron 70B Instruct HF is a language model from nvidia with a 32K-token context window. Provider list price is $0.88 per million input tokens and $0.88 per million output; on Allocate you pay $0.94 and $0.94 with the 7% transaction fee. The weights are open under Llama community, so you can fine-tune it and own the result.

## Pricing

| | Provider list | On Allocate |
| --- | --- | --- |
| Input, per M tokens | $0.88 | $0.94 |
| Output, per M tokens | $0.88 | $0.94 |

Token usage bills at the provider list price plus the 7% transaction fee. Prices checked 2026-07-08.

## Facts

| Field | Value |
| --- | --- |
| Lab | nvidia |
| Modality | Language |
| Context window | 32K tokens |
| License | Llama community |
| Open weights | Yes |
| Fine-tunable | Yes, on your data |
| Catalog id | nvidia/llama-3.1-nemotron-70b-instruct-hf |

## What a real workload costs

Take 1,000,000 requests a month at 1,200 input and 350 output tokens each: 1,200M input and 350M output tokens. At list prices that is 1,200 × $0.88 + 350 × $0.88 = $1,364 a month. Billed on Allocate it is $1,459 with the 7% transaction fee.

## Common questions

### How much does Llama 3.1 Nemotron 70B Instruct HF cost per million tokens?

Provider list price is $0.88 per million input tokens and $0.88 per million output tokens. On Allocate you pay list plus the 7% transaction fee: $0.94 in and $0.94 out.

### What context window does Llama 3.1 Nemotron 70B Instruct HF have?

32,768 tokens (32K). At roughly 0.75 words per token, that is about 25k words of English text per request.

### Can I fine-tune Llama 3.1 Nemotron 70B Instruct HF?

Yes. Llama 3.1 Nemotron 70B Instruct HF is an open-weights model under the Llama community license. Read the license terms before fine-tuning for commercial use. On Allocate the trained weights stay inside your boundary and belong to you.

### How do I call Llama 3.1 Nemotron 70B Instruct HF on Allocate?

Send nvidia/llama-3.1-nemotron-70b-instruct-hf in the model field of the OpenAI-compatible endpoint at api.allocate.network/v1, or point a route name (like prod/support-agent) at it so you can swap the model later without a deploy.

---

[HTML page](https://allocate.network/models/nvidia-llama-3-1-nemotron-70b-instruct-hf) · [Machine-readable catalog](https://allocate.network/catalog.json)