Skip to content
LLMs

Cheapest open-weight LLMs

Mistral: Mistral Nemo is the cheapest option at $0.019 per million input tokens. Models whose weights you can download and run yourself. The API price here is what you pay to avoid operating the hardware.

157 models qualify

Ranked by blended cost - input weighted 75%, output 25%.

Language models with price per million tokens and context window
#ModelBlended / 1MInput / 1MOutput / 1MContextCapabilities
1FreeFreeFree262K
Free tierReasoningToolsVisionOpen weights
2FreeFreeFree262K
Free tierReasoningToolsVisionOpen weights
3FreeFreeFree66K
Free tierReasoningToolsOpen weights
4FreeFreeFree256K
Free tierReasoningToolsOpen weights
5FreeFreeFree256K
Free tierReasoningToolsVisionOpen weights
6
Mistral AI logo
Mistral Nemo

Mistral AI

$0.022$0.019$0.030131K
ToolsOpen weights
7
inclusionAI logo
Ling 3.0 Flash

inclusionAI

$0.032$0.021$0.063262K
ReasoningToolsOpen weights
8$0.041$0.017$0.112131K
Open weights
9$0.042$0.040$0.0508K
Open weights
10$0.055$0.030$0.130131K
ReasoningToolsOpen weights
11
Mistral AI logo
Mistral Small 3

Mistral AI

$0.057$0.050$0.08033K
Open weights
12$0.057$0.050$0.080131K
ToolsOpen weights
13
Inference.net logo
Schematron V2 Turbo

Inference.net

$0.060$0.030$0.150128K
Open weights
14
MythoMax 13B logo
MythoMax 13B

MythoMax 13B

$0.060$0.060$0.0608K
Open weights
15$0.063$0.050$0.100131K
VisionOpen weights
16$0.070$0.037$0.170131K
ReasoningToolsOpen weights
17$0.071$0.027$0.20160K
Open weights
18$0.075$0.060$0.1201.3M
ReasoningToolsOpen weights
19$0.075$0.060$0.120262K
Free tierReasoningToolsOpen weights
20$0.075$0.050$0.150131K
ToolsVisionOpen weights
21$0.077$0.044$0.1778K
Open weights
22$0.084$0.048$0.193262K
ToolsOpen weights
23$0.087$0.050$0.200262K
ReasoningToolsOpen weights
24
Microsoft logo
Phi 4

Microsoft

$0.088$0.070$0.14016K
Open weights
25$0.090$0.060$0.180131K
Free tierReasoningToolsVisionOpen weights

Frequently asked

What is the cheapest LLM for self-hostable models?

Mistral: Mistral Nemo at $0.019 per million input tokens and $0.030 per million output tokens, with a 131K token context window.

How many models qualify for self-hostable models?

157 of the 340 models tracked here meet the criteria for this list.

How much do prices vary within this category?

By roughly 240×, from Mistral Nemo at the bottom to Kimi K3 at the top.