Skip to content
LLMs

Cheapest LLMs for agentic workloads

Mistral: Mistral Nemo is the cheapest option at $0.019 per million input tokens. Agents make many calls per task, so per-token price compounds fast. These models support tool calling and structured output, the two things an agent loop cannot run without.

252 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
1$-1000000.0000$-1000000.0000$-1000000.00002M
ReasoningToolsVision
2
OpenRouter logo
Auto Router

OpenRouter

$-1000000.0000$-1000000.0000$-1000000.00002M
ReasoningToolsVision
3FreeFreeFree262K
Free tierReasoningToolsVisionOpen weights
4FreeFreeFree262K
Free tierReasoningToolsVisionOpen weights
5FreeFreeFree512K
Free tierReasoningToolsVision
6FreeFreeFree66K
Free tierReasoningToolsOpen weights
7FreeFreeFree200K
Free tierReasoningToolsVision
8
Mistral AI logo
Mistral Nemo

Mistral AI

$0.022$0.019$0.030131K
ToolsOpen weights
9
inclusionAI logo
Ling 3.0 Flash

inclusionAI

$0.032$0.021$0.063262K
ReasoningToolsOpen weights
10$0.055$0.030$0.1301M
ReasoningToolsVision
11$0.055$0.030$0.130131K
ReasoningToolsOpen weights
12$0.057$0.050$0.080131K
ToolsOpen weights
13
Inception logo
Mercury 2.5

Inception

$0.068$0.040$0.150260K
ReasoningTools
14$0.070$0.037$0.170131K
ReasoningToolsOpen weights
15$0.075$0.060$0.1201.3M
ReasoningToolsOpen weights
16$0.075$0.050$0.150131K
ToolsVisionOpen weights
17$0.084$0.048$0.193262K
ToolsOpen weights
18$0.087$0.050$0.200262K
ReasoningToolsOpen weights
19$0.090$0.060$0.180131K
Free tierReasoningToolsVisionOpen weights
20$0.090$0.060$0.180262K
Free tierReasoningTools
21
Reka AI logo
Reka Edge

Reka AI

$0.100$0.100$0.10016K
ToolsVisionOpen weights
22$0.100$0.100$0.100131K
ToolsVisionOpen weights
23$0.106$0.075$0.200256K
ToolsVisionOpen weights
24$0.107$0.086$0.1711.0M
ReasoningToolsOpen weights
25
IBM Granite logo
Granite 4.2 8B

IBM Granite

$0.107$0.060$0.250131K
ReasoningToolsOpen weights

Frequently asked

What is the cheapest LLM for multi-step agents?

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 multi-step agents?

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

How much do prices vary within this category?

By roughly 3,100×, from Mistral Nemo at the bottom to GPT-5.4 Pro at the top.