Update Anthropic suspends Fable 5 access
Live LLM API price index

What does a token cost?

Every major model's API pricing tracked over time, so you can compare costs across providers and models.

291 Models tracked
50 Providers
$0.01/$0.03 Cheapest I/O /1M
Latest update
New model
Jun 12, 2026
Moonshot AI ships Kimi K2.7 Code at $1.71 I/O avg /1M.
Mid $0.95/$4.00 262K ctx
Recent activity
Claude Fable 5 released Launch Jun 2026
N
Nemotron 3 Ultra released Launch Jun 2026
Qwen3.7 Plus released Launch Jun 2026
M
MiniMax M3 released Launch May 2026
See the full timeline →

Price table


Filter by provider
1 provider selected

19 models of 291

Clear filters
Model Tier Input /1M Output /1M Cached in /1M Context I/O avg /1M Δ since launch Released
Mid $2.00 $6.00 $0.2 65K $3.00 Apr 2024
Mid $2.00 $6.00 $0.2 131K $3.00 Nov 2024
Mid $1.50 $7.50 128K $3.00 Apr 2026
Mid $0.4 $2.00 $0.04 131K $0.8 May 2025
Mid $0.4 $2.00 $0.04 131K $0.8 Aug 2025
Mid $0.4 $2.00 $0.04 262K $0.8 Dec 2025
Mid $0.5 $1.50 $0.05 262K $0.75 Dec 2025
Frontier $0.5 $1.50 262K $0.75 Dec 2025
Mid $0.3 $0.9 $0.03 256K $0.45 Aug 2025
Mid $0.351 $0.555 128K $0.402 Mar 2025
Mid $0.2 $0.6 $0.02 32K $0.3 Feb 2025
Mid $0.2 $0.2 $0.02 262K $0.2 Dec 2025
Mid $0.1 $0.3 $0.01 32K $0.15 Oct 2025
Mid $0.15 $0.15 $0.015 262K $0.15 Dec 2025
Small $0.1 $0.3 262K $0.15 decreased 42.9% Mar 2026
Mid $0.075 $0.2 128K $0.1062 Jun 2025
Mid $0.1 $0.1 $0.01 131K $0.1 Dec 2025
Mid $0.05 $0.08 32K $0.0575 Jan 2025
Mid $0.02 $0.03 131K $0.0225 Jul 2024

I/O avg assumes a 3:1 input-to-output token mix — a rough stand-in for a typical chat/agent workload, so models with different input vs. output pricing can be ranked on one number.