Home/Use Cases/Low Cost

Best LLM for Low Cost

12 models ranked for low cost tasks. Sorted by benchmark quality score, with price as a secondary factor.

Best Quality

Gemini 2.0 Flash

Google

ELO 1330

Cheapest Option

Mistral 7B

Mistral AI

$0.04/1M/1M input

All Models for Low Cost

#	Model	Provider	Input / 1M	Output / 1M	ELO	Flags
🥇	Gemini 2.0 Flash	Google	$0.10/1M	$0.40/1M	1330
🥈	DeepSeek V3	DeepSeek	$0.27/1M	$1.10/1M	1320	OSS
🥉	Qwen 2.5 72B	Alibaba	$0.35/1M	$0.40/1M	1280	OSS
4	GPT-4o mini	OpenAI	$0.15/1M	$0.60/1M	1272
5	Llama 3.1 70B	Meta	$0.35/1M	$0.40/1M	1247	OSS
6	Gemma 2 27B	Google DeepMind	$0.27/1M	$0.27/1M	1220	OSS
7	Gemini 1.5 Flash	Google	$0.07/1M	$0.30/1M	1211
8	Command R	Cohere	$0.15/1M	$0.60/1M	1200
9	Claude 3 Haiku	Anthropic	$0.25/1M	$1.25/1M	1179
10	Llama 3.1 8B	Meta	$0.06/1M	$0.06/1M	1176	FreeOSS
11	Phi-3.5 Mini	Microsoft	$0.10/1M	$0.10/1M	1112	FreeOSS
12	Mistral 7B	Mistral AI	$0.04/1M	$0.04/1M	1072	FreeOSS

Why We Picked These Models

Gemini 2.0 Flash

$0.10/1M/1MELO 1330

Gemini 2. Latest-gen quality with Flash-tier pricing.

DeepSeek V3

$0.27/1M/1MELO 1320

DeepSeek V3 shocked the AI industry by achieving GPT-4o-level performance at a fraction of the cost. Frontier-level quality at 10x lower cost than GPT-4o.

Qwen 2.5 72B

$0.35/1M/1MELO 1280

Qwen 2. Best-in-class for Chinese/Japanese/Korean languages.

Compare Top Models

Gemini 2.0 Flash vs DeepSeek V3 Gemini 2.0 Flash vs Qwen 2.5 72B Gemini 2.0 Flash vs GPT-4o mini Gemini 2.0 Flash vs Llama 3.1 70B

Frequently Asked Questions

What is the best LLM for low cost?

Gemini 2.0 Flash by Google is rated as the best model for low cost with an ELO score of 1330. Latest-gen quality with Flash-tier pricing.

What is the cheapest LLM for low cost?

Mistral 7B is the most affordable option for low cost at $0.04/1M per 1M input tokens. It is also available for free.

Is there a free LLM for low cost?

Yes, Llama 3.1 8B and Phi-3.5 Mini and Mistral 7B are available for free and suitable for low cost.