# DeepSeek-R1-Distill-Llama-70B

> R1 reasoning distilled onto a Llama 3.3 70B base — the largest of DeepSeek's January 2025 distill set.

- Maker: DeepSeek
- Line: DeepSeek R1 Distill
- Canonical page: https://ai-tldr.dev/models/deepseek-r1-distill-llama-70b/
- Updated: 2026-06-18

## In short

DeepSeek-R1-Distill-Llama-70B is an open-weight reasoning model that DeepSeek released on January 20, 2025 alongside its flagship DeepSeek-R1. Instead of being a from-scratch model, it takes Meta's Llama-3.3-70B-Instruct as a base and fine-tunes it on roughly 800K reasoning samples generated by the full DeepSeek-R1. It has a 128K-token context window and costs $0.80 per million input tokens and $0.80 per million output tokens. You can use it through OpenRouter and Groq (deprecated/decommissioned).

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 128K tokens (max output 32,768 tokens (recommended max generation length)) | [1] |
| Input price | $0.80 / 1M tokens | [2] |
| Output price | $0.80 / 1M tokens | [2] |
| Licence | MIT (base model derived from Llama-3.3-70B-Instruct, originally under the Llama 3.3 Community License) · open weights | [1] |
| Released | 20 January 2025 | [3] |
| Parameters | 70.6B (dense) | [1] |
| Knowledge cutoff | Not separately disclosed by DeepSeek; inherits the Llama 3.3 base pretraining cutoff of December 2023 | [1] |

## Sources

1. [DeepSeek-R1-Distill-Llama-70B — Hugging Face model card](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B)
2. [OpenRouter — pricing & API](https://openrouter.ai/deepseek/deepseek-r1-distill-llama-70b)
3. [Groq — model deprecations notice](https://console.groq.com/docs/deprecations)

## Overview

DeepSeek-R1-Distill-Llama-70B is an open-weight reasoning model that DeepSeek released on January 20, 2025 alongside its flagship DeepSeek-R1. Instead of being a from-scratch model, it takes Meta's Llama-3.3-70B-Instruct as a base and fine-tunes it on roughly 800K reasoning samples generated by the full DeepSeek-R1. The result is a 70.6B dense model that produces R1-style chain-of-thought — thinking through a problem step by step before answering — while running on the same hardware and inference stacks already built for Llama 3.3 70B.

It is the largest of the six original DeepSeek-R1 distill checkpoints (the others target 1.5B / 7B / 14B / 32B Qwen bases and an 8B Llama base). DeepSeek's own evaluation table puts the Llama-70B distill at the top of that set on math and reasoning, scoring 70.0 on AIME 2024, 94.5 on MATH-500, and 65.2 on GPQA Diamond — competitive with much larger reasoning systems while being downloadable and self-hostable.

The model is published under the MIT license, which permits commercial use, modification, and further distillation; the underlying Llama 3.3 weights remain subject to Meta's Llama 3.3 Community License. After launch it was widely served by inference providers — Groq, Together, Fireworks, OpenRouter and others — though hosted availability has since narrowed: Groq, for example, deprecated the model in September 2025 and decommissioned it in February 2026, pointing users to newer alternatives. The weights themselves remain available on Hugging Face.

## Benchmarks

- AIME 2024 (pass@1): 70% ([source](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B))
- MATH-500 (pass@1): 94.5% ([source](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B))
- GPQA Diamond (pass@1): 65.2% ([source](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B))
- LiveCodeBench (pass@1): 57.5% ([source](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B))

## Pricing

- Input: $0.80 / 1M tokens
- Output: $0.80 / 1M tokens

Open weights, so cost varies by host; example list price from OpenRouter. Free to self-host.

Source: https://openrouter.ai/deepseek/deepseek-r1-distill-llama-70b

## Strengths

- Strong math and step-by-step reasoning for its size: 94.5 on MATH-500 and 70.0 on AIME 2024 (pass@1) in DeepSeek's own evaluation
- Drops into existing Llama 3.3 70B infrastructure — same architecture, tokenizer family, and serving stacks (vLLM, SGLang, TGI, Ollama, llama.cpp)
- Open weights under a permissive MIT license, so it can be self-hosted, quantized, and further fine-tuned or distilled without API lock-in
- Exposes its chain-of-thought between <think> tags, which is useful for debugging reasoning and for building transparent agent loops
- Quantizes well to 4-bit, putting a 70B reasoning model within reach of a single high-memory GPU or a multi-GPU workstation

## Best for

- Self-hosted reasoning assistant for math, logic, and structured problem-solving where you want to keep data on-prem
- Code generation and debugging that benefits from explicit step-by-step thinking (57.5 on LiveCodeBench pass@1)
- A drop-in reasoning upgrade for teams already serving Llama 3.3 70B who want chain-of-thought without changing their stack
- Generating reasoning traces / synthetic data to further distill smaller student models
- Research and evaluation of open reasoning models under a permissive license

## How to access

- [OpenRouter](https://openrouter.ai/deepseek/deepseek-r1-distill-llama-70b) — model id `deepseek/deepseek-r1-distill-llama-70b`
- [Groq (deprecated/decommissioned)](https://console.groq.com/docs/model/deepseek-r1-distill-llama-70b) — model id `deepseek-r1-distill-llama-70b`

## DeepSeek R1 Distill versions

- [DeepSeek-R1-0528-Qwen3-8B](https://ai-tldr.dev/models/deepseek-r1-0528-qwen3-8b/) (29 May 2025)
- DeepSeek-R1-Distill-Llama-70B (20 January 2025) — this page
- [DeepSeek-R1-Distill-Qwen-32B](https://ai-tldr.dev/models/deepseek-r1-distill-qwen-32b/) (20 January 2025)
- [DeepSeek-R1-Distill-Qwen-14B](https://ai-tldr.dev/models/deepseek-r1-distill-qwen-14b/) (20 January 2025)
- [DeepSeek-R1-Distill-Llama-8B](https://ai-tldr.dev/models/deepseek-r1-distill-llama-8b/) (20 January 2025)
- [DeepSeek-R1-Distill-Qwen-7B](https://ai-tldr.dev/models/deepseek-r1-distill-qwen-7b/) (20 January 2025)
- [DeepSeek-R1-Distill-Qwen-1.5B](https://ai-tldr.dev/models/deepseek-r1-distill-qwen-1-5b/) (20 January 2025)

## FAQ

### Is DeepSeek-R1-Distill-Llama-70B the same as DeepSeek-R1?

No. DeepSeek-R1 is the full 671B-parameter Mixture-of-Experts reasoning model. The Llama-70B distill is a separate, much smaller 70.6B dense model: it takes Meta's Llama-3.3-70B-Instruct and fine-tunes it on reasoning samples generated by DeepSeek-R1. It mimics R1's chain-of-thought style at a fraction of the size, but it is not R1 itself.

### What license is it released under, and can I use it commercially?

DeepSeek published the distill weights under the MIT license, which permits commercial use, modification, and redistribution. Because the model is derived from Llama-3.3-70B-Instruct, the underlying base weights are also subject to Meta's Llama 3.3 Community License, so review both before deploying.

### What hardware do I need to run it?

It is a 70.6B dense model, so full-precision serving needs roughly 140GB+ of GPU memory (multiple GPUs). With 4-bit quantization it can fit on a single high-memory GPU or a multi-GPU workstation, and it runs in standard Llama 3.3 stacks such as vLLM, SGLang, Ollama, and llama.cpp.

### Is it still available through hosted APIs?

The weights remain on Hugging Face and can be self-hosted indefinitely. Hosted API access has narrowed over time, though — Groq, an early host, deprecated the model in September 2025 and decommissioned it in February 2026, directing users to newer models. Check each provider's current catalog before relying on it.

## Links

- [DeepSeek-R1-Distill-Llama-70B — Hugging Face model card](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B)
- [DeepSeek-R1 — official GitHub repository](https://github.com/deepseek-ai/DeepSeek-R1)
- [OpenRouter — pricing & API](https://openrouter.ai/deepseek/deepseek-r1-distill-llama-70b)
- [Groq — model docs](https://console.groq.com/docs/model/deepseek-r1-distill-llama-70b)
- [Groq — model deprecations notice](https://console.groq.com/docs/deprecations)
- [Artificial Analysis — context window & price analysis](https://artificialanalysis.ai/models/deepseek-r1-distill-llama-70b)
- [Meta Llama 3.3 70B — base model card (Dec 2023 cutoff)](https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/deepseek-r1-distill-llama-70b/