Skip to content

Cheapest LLMs by Price per Token

The cheapest LLMs are the large language models with the lowest API cost per token. Pricing is quoted per million (1M) tokens, with output tokens usually costing more than input. The table below ranks models from lowest to highest output price — a good starting point for high-volume, cost-sensitive workloads where you can trade some quality for price.

Ranked list (top 25)

#ModelInput / 1MOutput / 1MContext
1DeepSeek-OCR
NovitaAIOpen weights
$0.03$0.038K
2deepseek/deepseek-ocr-2
NovitaAIOpen weights
$0.03$0.038K
3baichuan-m2-32b
NovitaAIOpen weights
$0.07$0.07131K
4DeepSeek R1 0528 Qwen3 8B
NovitaAIOpen weights
$0.06$0.09128K
5DeepSeek R1 Distill Llama 70B
Helicone
$0.03$0.13128K
6AutoGLM-Phone-9B-Multilingual
NovitaAIOpen weights
$0.04$0.1466K
7Amazon Nova Micro 1.0
NanoGPT
$0.04$0.14128K
8Amazon: Nova Micro 1.0
Kilo Gateway
$0.04$0.14128K
9Amazon: Nova Micro 1.0
Amazon
$0.04$0.14128K
10DeepSeek R1 Distill Llama 70B
FastRouterOpen weights
$0.03$0.14131K
11DeepSeek R1 Distill Qwen 7B
Alibaba (China)
$0.07$0.1433K
12Arcee AI: Trinity Mini
Arcee AI
$0.05$0.15131K
13Arcee AI: Trinity Mini
Kilo GatewayOpen weights
$0.05$0.15131K
14Cohere: Command R7B (12-2024)
Kilo GatewayOpen weights
$0.04$0.15128K
15Cohere: Command R7B (12-2024)
CohereOpen weights
$0.04$0.15128K
16Command R7B
OpenRouterOpen weights
$0.04$0.15128K
17Command R7B
Merge GatewayOpen weights
$0.04$0.15128K
18Command R7B Arabic
CohereOpen weights
$0.04$0.15128K
19DeepSeek R1 Distill Qwen 14B
NovitaAIOpen weights
$0.15$0.1533K
20Baichuan 4 Air
NanoGPT
$0.16$0.1633K
21Arcee AI: Spotlight
Kilo GatewayOpen weights
$0.18$0.18131K
22DeepSeek V4 Flash
OpenRouterOpen weights
$0.09$0.181M
23ByteDance: UI-TARS 7B
ByteDance
$0.10$0.20128K
24ByteDance: UI-TARS 7B
Kilo Gateway
$0.10$0.20128K
25Coding MiniMax M2.7
AIHubMixOpen weights
$0.20$0.20205K

Prices are per 1M tokens (USD); confirm with the provider. Updated regularly.

More LLM rankings

Frequently asked questions