LLM Price Comparison

Input, output and cache pricing for every model, sorted per 1M tokens with filters.

FreeNo loginData from OpenRouter

Showing 458 of 464 models

Space Bunny Alpha

stealth

FreeFreeFree

Ling 3.0 Flash Sante (free)

inclusionAI

FreeFreeFree

Qwen3.8 27B (free)

Qwen

FreeFreeFree

Dots3-Note Preview (free)

Dots Studio

FreeFreeFree

LFM2.5-2.6B (free)

LiquidAI

FreeFreeFree

Nemotron 3.5 Lightning (free)

NVIDIA

FreeFreeFree

Inkling Small (free)

Thinking Machines

FreeFreeFree

Laguna S 2.1 (free)

Poolside

FreeFreeFree

Inkling (free)

Thinking Machines

FreeFreeFree

Laguna XS 2.1 (free)

Poolside

FreeFreeFree

North Mini Code (free)

Cohere

FreeFreeFree

Nemotron 3.5 Content Safety (free)

NVIDIA

FreeFreeFree

Nemotron 3 Ultra (free)

NVIDIA

FreeFreeFree

Nemotron 3 Nano Omni (free)

NVIDIA

FreeFreeFree

Gemma 4 26B A4B (free)

Google

FreeFreeFree

Gemma 4 31B (free)

Google

FreeFreeFree

Lyria 3 Pro Preview

Google

FreeFreeFree

Lyria 3 Clip Preview

Google

FreeFreeFree

Nemotron 3 Super (free)

NVIDIA

FreeFreeFree

Free Models Router

openrouter

FreeFreeFree

Mistral Nemo

Mistral

$0.019$0.030$0.022

Ling 3.0 Flash VL

inclusionAI

$0.021$0.062$0.031

Ling 3.0 Flash

inclusionAI

$0.021$0.063$0.032

gpt-oss-20b

OpenAI

$0.018$0.090$0.036

DeepSeek V4 Flash Latest

DeepSeek

$0.010$0.131$0.040

Granite 4.0 Micro

IBM

$0.017$0.112$0.041

Llama 3 8B Lunaris

Sao10K

$0.040$0.050$0.042

Nex-N2.5-Mini

Nex AGI

$0.025$0.100$0.044

gpt-oss-20b (batch)

OpenAI

$0.024$0.112$0.046

Qwen3.7 Flash

Qwen

$0.030$0.130$0.055

gpt-oss-120b (batch)

OpenAI

$0.030$0.136$0.056

Mistral Small 3

Mistral

$0.050$0.080$0.057

Llama 3.1 8B Instruct

Meta

$0.050$0.080$0.057

Schematron V2 Turbo

Inference.net

$0.030$0.150$0.060

Nova Micro 1.0

Amazon

$0.035$0.140$0.061

Gemma 3 4B

Google

$0.050$0.100$0.063

Command R7B (12-2024)

Cohere

$0.037$0.150$0.066

Mercury 2.5

Inception

$0.040$0.150$0.068

GPT-5 Nano (batch)

OpenAI

$0.025$0.200$0.069

gpt-oss-120b

OpenAI

$0.037$0.170$0.070

Llama 3.2 1B Instruct

Meta

$0.027$0.201$0.071

Laguna XS 2.1

Poolside

$0.060$0.120$0.075

Ministral 3 8B 2512 (batch)

Mistral

$0.075$0.075$0.075

Gemma 3 12B

Google

$0.050$0.150$0.075

GLM Flash Latest

Z.ai

$0.020$0.248$0.077

Hy-MT2-1.8B

Tencent

$0.044$0.177$0.077

Qwen3 30B A3B Instruct 2507

Qwen

$0.048$0.193$0.084

Nemotron 3.5 Lightning

NVIDIA

$0.060$0.160$0.085

Solar Mini 4

Upstage

$0.050$0.200$0.087

Nemotron 3 Nano 30B A3B

NVIDIA

$0.050$0.200$0.087

Gemini 2.5 Flash Lite (batch)

Google

$0.050$0.200$0.087

GPT-4.1 Nano (batch)

OpenAI

$0.050$0.200$0.087

MythoMax 13B

gryphe

$0.080$0.110$0.087

Phi 4

Microsoft

$0.070$0.140$0.088

Ling 3.0 Flash Fin

inclusionAI

$0.060$0.180$0.090

Schematron V2 Small

Inference.net

$0.050$0.230$0.095

GLM 5.3 Flash (batch)

Z.ai

$0.060$0.200$0.095

Reka Edge

rekaai

$0.100$0.100$0.100

Ministral 3 3B 2512

Mistral

$0.100$0.100$0.100

GPT-6 Luna Pro (batch)

OpenAI

$0.050$0.250$0.100

Cost estimator

Enter one task's usage to convert pricing into per-call and monthly cost.

No model selected yet. Use the estimate button on a row to add one.

Prices are USD per 1M tokens, converted from OpenRouter's official rates. The cache price applies to reused long prompts: a cache hit bills the input at the discounted rate.

Input, output and cache pricing for every model, sorted per 1M tokens with filters.

Every model on OpenRouter bills input and output separately, and the gap between the two is routinely four or five times. Sorting on output alone picks the wrong model; so does sorting on input alone. This table puts both prices, the cache-read rate, context length and capability tags on one row so you can match a model to your own usage shape — and the estimator underneath converts pricing into what one task, and one month, actually cost.

How to use

  1. 1Narrow the list with the search box or the vendor dropdown, then stack filters for price band, context length and capabilities.
  2. 2Click a column header to sort by input, output or blended price — blended weights input three to one, which tracks a real bill more closely.
  3. 3Enter one task's input and output tokens in the estimator to compare monthly cost across candidates.

FAQ