DeepSeek R1 Distill Llama 70B

A 70B reasoning model distilled from DeepSeek R1 onto Llama 3.3 70B, with weights under the MIT license.

Params: 70.6B
Context: 131K tokens
Max output: 33K
Released: Jan 2025
License: MIT

Estimated cost to self-host

DeepSeek R1 Distill Llama 70B needs about 46 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 4× RTX 5060 Ti at about $403 a month.

Single node ≤ 8

What it costs to self-host DeepSeek R1 Distill Llama 70B at each precision: the memory it needs, the cheapest rental that holds it, that rental's month and the API volume it pays off at
Precision	Memory	Cheapest fit (32K context)(Full 131K context)	Cost /mo
4-bit INT4 / FP4	46 GB84 GB	4× RTX 5060 Ti	$403$720
8-bit FP8 / INT8	83 GB94 GB	4× RTX 3090	$720$806
16-bit FP16 / BF16	154 GB186 GB	8× RTX 3090	$1,440$3,398

Estimates based on median on-demand rates for Nvidia GPUs. Memory is weights plus KV cache for one request, using FP8 KV cache where supported and FP16/BF16 otherwise. No guarantee of runtime support, usable performance, or that a matching quantized build exists. Pricing methodology.

GPUhub

Fully Dedicated GPU Servers

No Virtualization. No Shared GPUs

Stable Performance for AI Training & Inference

Full Control for Production Workloads

Learn more

Capabilities

What DeepSeek R1 Distill Llama 70B accepts and can do, as published by DeepSeek.

Text Accepts and generates natural-language text.
Reasoning Thinks before it answers, always on or as a switchable mode.

Common questions

What is DeepSeek R1 Distill Llama 70B good for?

Reasoning work on a dense 70B model. MIT-licensed weights, far lighter to self-host than the 671B R1 it was distilled from, on a single high-memory GPU.

When is DeepSeek R1 Distill Llama 70B not a good fit?

Tool use and image input: it has no function calling or structured output, and takes text only.

What is the cheapest way to run DeepSeek R1 Distill Llama 70B?

No provider we track hosts it; self-hosting on the cheapest rental that fits (4x RTX 5060 Ti, $0.56 per hour) costs about $403 a month.

Can I self-host DeepSeek R1 Distill Llama 70B?

Yes. The weights are MIT licensed. At 4-bit it needs about 46 GB of GPU memory, which starts at roughly $403 a month on the cheapest rental that fits.