01
Input and output token prices
LLM APIs usually charge separately for tokens sent to the model and tokens generated in the response. LLM Cost normalizes both rates per one million tokens so providers are easier to compare.
LLM Cost
LLM API pricing, without the guesswork.
Input and output prices for every major LLM. Browse by provider, compare two models, or estimate a run cost in seconds.
Cheapest paid today · output $0.020/M · input $0.010/M
Models1536
Providers88
Live pricing
Compare LLM API prices
A current cross-provider snapshot, grouped by comparable capability so flagship, balanced, and budget models are not presented as direct substitutes.
TierModelInput / 1MOutput / 1MContext
Frontier
GPT-6 Sol
OpenAI
Input / 1M
$2.00
Output / 1M
$10.00
Context922K
Frontier
DeepSeek V4 Pro 0813
Deepseek
Input / 1M
$0.346
Output / 1M
$1.04
Context1M
Balanced
DeepSeek V4.1 Flash
Deepseek
Input / 1M
$0.200
Output / 1M
$0.650
Context1M
Balanced
Gemini 3.8 Flash
Input / 1M
$0.750
Output / 1M
$3.75
Context1M
Budget
GPT-6 Luna
OpenAI
Input / 1M
$0.100
Output / 1M
$0.500
Context922K
Budget
Gemini 3.5 Flash-Lite
Input / 1M
$0.300
Output / 1M
$2.50
Context1M
Capability tiers describe product positioning, not an independent quality benchmark. Open a model to inspect every available provider price.
Estimate
Pricing calculator
Select a provider and model, enter your token counts, and get an instant cost estimate.
Configure
Step 4 of 4 · Tokens
Provider
Family
Model
Listed rates
GPT-6 Sol
Estimated run
$0.0070
From 1,000 in · 500 out
Input / 1M
$2.00
Output / 1M
$10.00
Prices are official list rates per million tokens.
Full pricing detailLLM pricing basics
How to calculate LLM API cost
Compare current LLM API prices using the same unit, then estimate the cost for your actual input and output token volume.
02
Estimate the real request cost
Multiply input tokens by the input rate, add output tokens multiplied by the output rate, then divide by one million. The calculator below applies that formula to the live catalog.
03
Compare the same model across providers
A model can have different prices across its native API and inference providers. Model and comparison pages show the current provider matrix instead of treating one list price as universal.
Prices sync from the catalog throughout the day. Browse the LLM pricing directory or open the calculator below for a workload estimate.
Value
Cheapest LLM APIs right now
The 5 most affordable broadly available language models by output price.
#1 · MetaLlama 3.2 1B Instruct
Meta
Out / 1M
$0.020
#2 · MetaLlama 3.2 3B Instruct
Meta
Out / 1M
$0.020
#3 · MetaDeepseek R1 Distill Llama 8b
Meta
Out / 1M
$0.025
#4 · MetaLlama3 2 3b
Meta
Out / 1M
$0.025
#5 · MetaLlama 3.1 8B
Meta
Out / 1M
$0.030
Showing 5 of 402 broadly available models · sort and filter on the models page
Popular comparisons
Click any pair to see their pricing and specs side by side.
- AMistral LargeVSBOpen Mistral 7B ($2.00 vs $0.250/1M)Open compareView →
- AMixtral 8x22BVSBOpen Mistral 7B ($1.20 vs $0.250/1M)Open compareView →
- AAutoglm Phone 9BVSBGLM 4.7 ($0.035 vs $0.600/1M)Open compareView →
- AAutoglm Phone 9BVSBGLM 4.5 Air ($0.035 vs $0.200/1M)Open compareView →
- AClaude Sonnet 4.6VSBQwen2.5 14B ($3.00 vs $0.200/1M)Open compareView →
- ADeepSeekVSBQwen3 14B ($0.280 vs $0.050/1M)Open compareView →
- AGPT-4o miniVSBGPT-5 Nano ($0.150 vs $0.050/1M)Open compareView →
- AClaude Sonnet 4.6VSBDeepSeek R1 0528 Tput ($3.00 vs $0.550/1M)Open compareView →
Insights
Market overview
See how top models compare on price and where the market clusters.
Top model prices
Input vs. output cost per 1M tokens
Price distribution
Number of models at each price range ($/M)
What you get
One place for LLM prices
Browse the full model catalog, see who's cheapest, and compare two models side by side. All in one place.
Model directory
Browse models from multiple providers, filter by cost, and check input and output pricing per million tokens.
GPT-6 Sol
OpenAI
out $10.00Claude Fable 5.1
Anthropic
out $50.00DeepSeek V4.1 Flash
Deepseek
out $0.650
Multi-provider coverage
Prices from all major providers in one place: Anthropic, OpenAI, Google, Meta, and more.
Live index
1536models88 providers
List $/1MSort & filterSame schema
- Anthropic
- OpenAI
- Meta
- Mistral
- DeepSeek
Side-by-side compare
Compare the cost and specs of any two models side by side, with charts to visualize the difference.
Mistral Large
Mistral AI
In / 1M$2.00
Out / 1M$6.00
vs
Open Mistral 7b
Mistral AI
In / 1M$0.250
Out / 1M$0.250
Pricing gridSpec diffScatter charts
Live catalog preview: Mistral Large · Open Mistral 7B ($2.00 vs $0.250/1M)
Ready to compare LLM pricing?
Browse the full catalog, compare two models, or estimate a run cost.