AI Model Benchmarks

Real-Time AI Model Performance Data

Compare models, track usage trends, and make data-driven decisions with comprehensive benchmarks from millions of API calls

0.0T
Tokens Processed
0+
Models Tracked
0M
Tool Calls
0.0M
Images Processed

Model Leaderboard

Top performing models by token usage

1

Grok Code Fast 1

by x-ai

12%
1.04T
tokens this week
Programming
Fast inference
Code completion
Multi-language
2

Grok 4 Fast (free)

by x-ai

1293%
854B
tokens this week
General
Free tier
General purpose
Fast response
3

Claude Sonnet 4

by anthropic

6%
558B
tokens this week
General
Advanced reasoning
Long context
Safety focused
4

Gemini 2.5 Flash

by google

6%
344B
tokens this week
General
Multimodal
Fast
Cost effective
5

DeepSeek V3.1 (free)

by deepseek

17%
204B
tokens this week
Programming
Open source
Code focused
Free
6

Gemini 2.0 Flash

by google

10%
167B
tokens this week
General
Latest version
Improved speed
Better accuracy
7

GPT-5

by openai

23%
145B
tokens this week
General
Most advanced
Multi-task
Enterprise ready
8

DeepSeek V3 0324

by deepseek

1%
142B
tokens this week
Programming
Specialized
High accuracy
Developer focused

Market Analysis

Understanding AI model adoption and usage patterns

Market Share by Provider

google41.1% • 41.1B
openai21.1% • 40.1B
anthropic20% • 38.2B
mistralai8.7% • 16.5B
meta-llama8.3% • 15.7B
others0.8% • 1.4B

Usage by Category

Usage Trends

Last 14 days of activity from real site data

00111
Total
Reactions
Shares
Loading…

Performance Metrics

Key performance indicators across models

Average Response Time

45ms
Fast Models
120ms
Standard
350ms
Complex

Cost Efficiency

GPT-5$0.03/1K tokens
Claude Sonnet 4$0.015/1K tokens
Gemini 2.5 Flash$0.001/1K tokens

Community Contributions

Share your use cases, request new benchmarks, and help the community make informed decisions

Share Use Cases

Submit your real-world implementations and success stories

Request Benchmarks

Suggest new models or metrics you'd like to see compared

Contribute Data

Help improve our benchmarks with your performance data

0+
Use Cases Shared
0
Models Tracked
0.0K+
Community Members

Start Building with the Right Model

Use our benchmarks to make informed decisions and optimize your AI implementations