What is DeepSeek R1 good for?

Math and code tasks that gain from step-by-step reasoning. MIT-licensed weights allow self-hosting or fine-tuning, on a multi-GPU node.

When is DeepSeek R1 not a good fit?

Cost-sensitive or multimodal work: it is text only and sits at the top of DeepSeek's price range. DeepSeek's later hybrid models fold R1-style reasoning into a single chat model.

What is the cheapest way to run DeepSeek R1?

Hosted, unless you push serious volume. Novita charges $0.70 in / $2.50 out per 1M tokens. The cheapest rental that fits is 8x A100 at $10,138 a month, which costs the same as about 10B tokens a month on that API.

Can I self-host DeepSeek R1?

Yes. The weights are MIT licensed. At 4-bit it needs about 373 GB of GPU memory, which starts at roughly $10,138 a month on the cheapest rental that fits.

Open weights · MIT

DeepSeek R1

DeepSeek's 671B MoE reasoning model (37B active) with chain-of-thought reasoning for math and code.

Jump to: Specs Pricing Self-hosting Similar models FAQ More from DeepSeek

Visit website Compare vs other LLMs →

Cheapest via Novita

Input: $0.70 /1M tokens
Output: $2.50 /1M tokens

3 providers

Key Specifications

Context window: 128K tokens
Released: Jan 2025
Parameters: 671B, 37B active
Inputs: Text

Capabilities Show details: Reasoning Thinks before it answers, always on or as a switchable mode. Function calling Connect to external tools, APIs, and systems. Structured output Return responses in structured formats like JSON.

Hosted API pricing

Input tokens M Output tokens M

Provider	Input /1M tokens	Output /1M tokens	Cached input /1M	Cost at 10M in + 2M out
Novita Cheapest	$0.70	$2.50	$0.35	$12.00	View
Azure	$1.35	$5.40		$24.30	View
Replicate	$3.75	$10.00		$57.50	View

Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.

Estimated cost to self-host

DeepSeek R1 needs about 373 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 8× A100 at about $10,138 a month.

That costs the same as roughly 10B tokens a month on Novita's API. Self-hosting is more expensive below that volume.

Single node ≤ 8

Precision	Memory	Cheapest fit (32K context)(Full 128K context)	Cost /mo	Break-even vs API
4-bitINT4 / FP4	373 GB380 GB	8× A100	$10,138$10,138	10B tokens /mo10B tokens /mo
8-bitFP8 / INT8	674 GB678 GB	8× RTX PRO 6000	$12,672$12,672	13B tokens /mo13B tokens /mo
16-bitFP16 / BF16	1,346 GB1,353 GB	8× B300	$45,446$45,446	45B tokens /mo45B tokens /mo

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. Break-even assumes a 5:1 input-to-output ratio. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Similarly priced models

The models nearest DeepSeek R1 by blended rate, each at its own cheapest provider.

Model	Blended /1M	Input /1M	Output /1M	Context	Cutoff	vs DeepSeek R1
GLM-4.7 Z.AI	$0.8667	$0.60	$2.20	205K		−13%
Gemini 3 Flash Preview Google Cloud	$0.9167	$0.50	$3.00	1M	Jan 2025	−8%
Qwen3.6-Plus Alibaba Cloud	$0.9167	$0.50	$3.00	1M		−8%
GLM-5.2 Z.AI	$0.95	$0.70	$2.20	1M		−5%
Nemotron 3 Ultra 550B A55B Nvidia	$0.99	$0.66	$2.64	1M	Sep 2025	−1%
DeepSeek R1 This model DeepSeek	$1.00	$0.70	$2.50	128K
DeepSeek R1 Turbo DeepSeek	$1.00	$0.70	$2.50	64K		0%
Llama 3 70B Meta	$1.00	$0.65	$2.75	8K	Dec 2023	0%
Qwen3-VL-235B-A22B Thinking Alibaba Cloud	$1.00	$0.40	$4.00	262K		0%
Kimi K2.6 Moonshot AI	$1.0333	$0.60	$3.20	262K		+3%
Grok Build 0.1 xAI	$1.1667	$1.00	$2.00	256K		+17%

Prices are USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.

Packet·ai

B200 @ $3.75/h on demand

No contract, no spot interrupts, no cluster minimums

Full root SSH, web dashboard, CI/CD ready

Reserve Now

Frequently Asked Questions

What is DeepSeek R1 good for?: Math and code tasks that gain from step-by-step reasoning. MIT-licensed weights allow self-hosting or fine-tuning, on a multi-GPU node.
When is DeepSeek R1 not a good fit?: Cost-sensitive or multimodal work: it is text only and sits at the top of DeepSeek's price range. DeepSeek's later hybrid models fold R1-style reasoning into a single chat model.
What is the cheapest way to run DeepSeek R1?: Hosted, unless you push serious volume. Novita charges $0.70 in / $2.50 out per 1M tokens. The cheapest rental that fits is 8x A100 at $10,138 a month, which costs the same as about 10B tokens a month on that API.
Can I self-host DeepSeek R1?: Yes. The weights are MIT licensed. At 4-bit it needs about 373 GB of GPU memory, which starts at roughly $10,138 a month on the cheapest rental that fits.

More from DeepSeek

Model	Context	Input /1M	Output /1M
DeepSeek R1 Turbo Open weights	64K	$0.70	$2.50
DeepSeek R1 Distill Llama 70B Open weights	131K	$0.80	$0.80
DeepSeek V4 Flash Vision Exp	1M	$0.44	$1.32
DeepSeek V3 Turbo Open weights	64K	$0.40	$1.30
DeepSeek V4 Pro Open weights	1M	$0.45	$0.89
DeepSeek V4.1 Flash Open weights	1M	$0.30	$1.20
DeepSeek V3 Open weights	128K	$0.27	$1.12
DeepSeek V3.1 Open weights	131K	$0.27	$1.00