AI Model Comparison
Credit System Overview
Getting Started
New users receive 100,000 initial credits to explore our services.
Credit Calculation
Credits are calculated per cell, with costs varying based on the model used.
OpenAI Language Models
Gpt35 Base
2 credits
Context:
16K tokens
Speed:
40 tokens/s
Performance:
59/100
Best For:
General purpose, cost-effective tasks
Gpt35 Turbo 16k
2 credits
Context:
16K tokens
Speed:
40 tokens/s
Performance:
59/100
Best For:
Longer context requirements
Gpt35 Turbo Instruct
2 credits
Context:
4K tokens
Speed:
45 tokens/s
Performance:
58/100
Best For:
Instruction-following tasks
Gpt4
914 credits
Context:
128K tokens
Speed:
27.9 tokens/s
Performance:
94/100
Best For:
Complex reasoning, high-accuracy requirements
Gpt4o
6 credits
Context:
128K tokens
Speed:
83.8 tokens/s
Performance:
100/100
Best For:
High-performance quick responses
Gpt4o Mini
3 credits
Context:
128K tokens
Speed:
80 tokens/s
Performance:
85/100
Best For:
Balanced performance and cost
Google Models
Gemini 15 Pro
5 credits
Context:
2M tokens
Speed:
58.3 tokens/s
Performance:
95/100
Best For:
Large context processing
Gemini 15 Ground
4 credits
Context:
1M tokens
Speed:
100 tokens/s
Performance:
90/100
Best For:
Balanced performance
Gemini 15 Flash
3 credits
Context:
128K tokens
Speed:
164.9 tokens/s
Performance:
84/100
Best For:
High-speed processing
Anthropic Models
Claude 35 Sonnet
190 credits
Context:
200K tokens
Speed:
78.6 tokens/s
Performance:
98/100
Best For:
Complex tasks with extensive context
Vision Models
Anthropic Vision
190 credits
Performance:
95/100
Best For:
Advanced image understanding and analysis
Capabilities:
- Advanced image understanding
- Detailed visual analysis
- Multi-modal processing
Gemini Vision
7 credits
Performance:
93/100
Best For:
Large-scale visual processing
Capabilities:
- Scene understanding
- Object detection
- Visual context analysis
Openai Vision
914 credits
Performance:
96/100
Best For:
Precise image analysis and description
Capabilities:
- High-accuracy recognition
- Context-aware analysis
- Detailed scene description
Image Generation Models
Dalle 2
4350 credits
Resolution:
Up to 1024x1024
Generation Time:
~4s
Performance:
87/100
Capabilities:
- Artistic interpretation
- Style variation
- Concept visualization
Dalle 3
8700 credits
Resolution:
Up to 2048x2048
Generation Time:
~6s
Performance:
94/100
Capabilities:
- Photorealistic output
- Complex compositions
- Precise details
Imagen 2
4350 credits
Resolution:
Up to 1024x1024
Generation Time:
~4s
Performance:
88/100
Capabilities:
- Photo-realistic output
- Style consistency
- Prompt accuracy
Imagen 3
8700 credits
Resolution:
Up to 2048x2048
Generation Time:
~5s
Performance:
92/100
Capabilities:
- Advanced composition
- Enhanced details
- Style control
Image Editing Models
Imagen 3 Edit
4350 credits
Resolution:
Up to 2048x2048
Editing Time:
~7s
Capabilities:
- Advanced editing
- Style transfer
- Seamless modifications
Model Comparison
| Model | Credits/Cell | Performance | Context Window | Best Use Case |
|---|---|---|---|---|
| Gpt35 Base | 2 | 59 | 16K tokens | General purpose, cost-effective tasks |
| Gpt35 Turbo 16k | 2 | 59 | 16K tokens | Longer context requirements |
| Gpt35 Turbo Instruct | 2 | 58 | 4K tokens | Instruction-following tasks |
| Gpt4 | 914 | 94 | 128K tokens | Complex reasoning, high-accuracy requirements |
| Gpt4o | 6 | 100 | 128K tokens | High-performance quick responses |
| Gpt4o Mini | 3 | 85 | 128K tokens | Balanced performance and cost |
| Gemini 15 Pro | 5 | 95 | 2M tokens | Large context processing |
| Gemini 15 Ground | 4 | 90 | 1M tokens | Balanced performance |
| Gemini 15 Flash | 3 | 84 | 128K tokens | High-speed processing |
| Claude 35 Sonnet | 190 | 98 | 200K tokens | Complex tasks with extensive context |
| Anthropic Vision | 190 | 95 | 200K tokens | Advanced image understanding and analysis |
| Gemini Vision | 7 | 93 | 2M tokens | Large-scale visual processing |
| Openai Vision | 914 | 96 | 128K tokens | Precise image analysis and description |
| Dalle 2 | 4350 | 87 | N/A | Creative image generation |
| Dalle 3 | 8700 | 94 | N/A | Professional image creation |
| Imagen 2 | 4350 | 88 | N/A | Standard image generation |
| Imagen 3 | 8700 | 92 | N/A | High-quality image generation |
| Imagen 3 Edit | 4350 | 90 | N/A | Professional image editing |
How to Choose the Right Model
Cost-Effective Tasks
Use GPT-3.5 or Gemini 1.5 Flash for:
- Simple text completion
- Basic analysis
- High-volume processing
Complex Analysis
Use GPT-4 or Claude for:
- Advanced reasoning
- Complex problem-solving
- High-accuracy requirements
Vision Tasks
Choose Vision Models for:
- Image analysis
- Visual content understanding
- Multi-modal applications
Image Generation
Select based on needs:
- DALL-E 3: Professional quality
- Imagen 3: Advanced editing
- DALL-E 2: Cost-effective creation
Usage Examples
Basic Analysis
100 cells using GPT-3.5
100 × 2 credits = 200 credits
Advanced Processing
50 cells using GPT-4
50 × 8 credits = 400 credits
Cost Comparison Chart
OpenAI Models
Google Models
Anthropic Models
Vision Models
Image Generation
Image Editing