AI Models

Compare 741 AI models across 13 providers. Pricing, context windows, and capabilities at a glance.

Last synced 12 hours ago

Estimate costs for your workload
Reset
Model Provider Input $/Mtok Output $/Mtok Context ↓ Capabilities Released
Llama 3 8B Instruct
OpenRouter $0.14 $0.14 8,192 open 2024-04-18
GPT-4 (older v0314)
OpenRouter $30 $60 8,191 tools 2023-05-28
GPT-4
OpenRouter $30 $60 8,191 tools 2023-11-06
Weaver (alpha)
OpenRouter $0.5 $0.75 8,000 N/A 2023-08-02
Inflection 3 Productivity
OpenRouter $2.5 $10 8,000 N/A 2024-10-11
Inflection 3 Pi
OpenRouter $2.5 $10 8,000 N/A 2024-10-11
ReMM SLERP 13B
OpenRouter $0.45 $0.65 6,144 open 2023-07-22
Seedream 4.5
OpenRouter $0 $0 4,096 vision, open 2025-12-23
ALLaM-2-7b
Groq $0 $0 4,096 open 2025-01-23
CodeLLaMa 7B Instruct Solidity
OpenRouter $0.8 $1.2 4,096 open 2025-04-14
GPT-3.5 Turbo Instruct
OpenRouter $1.5 $2 4,095 N/A 2023-09-28
GPT-3.5 Turbo (older v0613)
OpenRouter $1 $2 4,095 tools 2024-01-25
Canopy Labs Orpheus Arabic Saudi
beta
Groq N/A N/A 4,000 N/A 2025-12-16
Canopy Labs Orpheus V1 English
beta
Groq N/A N/A 4,000 N/A 2025-12-19
Mistral 7B Instruct v0.1
OpenRouter $0.11 $0.19 2,824 open 2023-09-28
Llama Prompt Guard 2 22M
beta
Groq $0.03 $0.03 512 open 2025-05-29
Prompt Guard 2 86M
beta
Groq $0.04 $0.04 512 open 2025-05-29
Whisper
Groq N/A N/A N/A open 2023-09-01
Whisper Large V3 Turbo
Groq N/A N/A N/A open 2024-10-01

About this AI model comparison

AI model pricing changes frequently as providers compete on cost and capability. This comparison covers models from Anthropic, OpenAI, Google, xAI, Mistral, Cohere, DeepSeek, and major inference aggregators including OpenRouter and Groq. Filter by capability to find the right fit: vision-capable models for image inputs, tool-calling models for agents, reasoning models for complex tasks, or open-weight models for self-hosting. Pricing is shown in dollars per million tokens; multiply by your expected token volume to estimate monthly costs.

Data synced daily from models.dev. Always verify pricing with the provider before deploying to production.