Guide
Models
Onsomble offers 35+ AI models. Learn which one to choose for different tasks.
Different AI models have different strengths. Some are faster and cheaper. Others are more capable but cost more. This guide helps you choose.
Choosing a Model
Quick Recommendations
| Use Case | Recommended Model | Why |
|---|---|---|
| Daily tasks | Claude Sonnet 4 | Great balance of speed, quality, cost |
| Complex analysis | Claude Opus 4.5 | Most capable reasoning |
| Quick questions | Claude Haiku 4.5 | Fast and cheap |
| Code tasks | GPT-4o or Claude Sonnet | Strong at coding |
| Long documents | Gemini 2.5 Pro | Large context window |
Factors to Consider
Quality — How good are the responses?
- Higher-tier models give better reasoning and nuance
- Lower-tier models work fine for simple tasks
Speed — How fast do you get responses?
- Smaller models respond faster
- Larger models take longer but may be worth the wait
Cost — How many credits per message?
- Ranges from 0.5 to 10 credits per message
- Complex queries use more credits regardless of model
Context Window — How much can the model “remember”?
- Larger windows handle longer conversations
- Important for documents with lots of content
Available Models
Anthropic (Claude)
| Model | Credits | Best For |
|---|---|---|
| Claude Opus 4.5 | 10 | Complex reasoning, difficult problems |
| Claude Sonnet 4.5 | 3 | Daily driver, great all-around |
| Claude Sonnet 4 | 2 | Balanced quality and cost |
| Claude Haiku 4.5 | 0.5 | Quick tasks, simple questions |
Anthropic strengths: Nuanced reasoning, following instructions, safety
OpenAI (GPT)
| Model | Credits | Best For |
|---|---|---|
| GPT-5 | 5 | Cutting-edge capability |
| GPT-5 Mini | 2 | Good balance |
| GPT-4o | 2 | Reliable all-around |
| GPT-4.1 Mini | 1 | Budget-friendly |
| o3 Mini | 1.5 | Reasoning tasks |
| o4 Mini | 2 | Advanced reasoning |
OpenAI strengths: Coding, broad knowledge, consistency
Google (Gemini)
| Model | Credits | Best For |
|---|---|---|
| Gemini 2.5 Pro | 3 | Long documents, analysis |
| Gemini 2.5 Flash | 1 | Fast responses |
| Gemini 3 Pro Preview | 4 | Latest capabilities |
Google strengths: Large context windows, multimodal understanding
Meta (Llama)
| Model | Credits | Best For |
|---|---|---|
| Llama 3.3 70B | 1 | Good open model |
| Llama 3.1 405B | 2 | Largest open model |
| Llama 3.1 70B | 1 | Balanced performance |
| Llama 3.1 8B | 0.5 | Quick and cheap |
Meta strengths: Open weights, good value, fast inference
DeepSeek
| Model | Credits | Best For |
|---|---|---|
| DeepSeek V3.2 | 1.5 | Reasoning and code |
| DeepSeek Chat V3.1 | 1 | General chat |
DeepSeek strengths: Strong reasoning, competitive pricing
xAI (Grok)
| Model | Credits | Best For |
|---|---|---|
| Grok 4.1 Fast | 2 | Quick responses |
| Grok 4 | 3 | Full capability |
| Grok 4 Fast | 1.5 | Balance of speed and quality |
xAI strengths: Real-time knowledge, direct responses
How to Change Models
In Notebook Chat
- Look at the bottom of the chat input
- Click the model name (shows current model)
- Browse or search for a model
- Select to switch
In General Chat
Same process — click the model selector and choose.
Model selection applies to future messages. Previous messages keep their original model.
Credit Costs Explained
Credits are consumed based on:
- Base model cost — Each model has a per-message rate
- Query complexity — Simple questions cost less than complex analysis
- RAG usage — Searching sources adds a small cost
- Web search — Optional, adds cost when enabled
Complexity Levels
| Level | Multiplier | Example Questions |
|---|---|---|
| Light | 1x | ”What is X?” / Simple lookups |
| Medium | 1.5x | ”Compare X and Y” / Analysis |
| Heavy | 2.5x | Multi-step research / Deep synthesis |
Example Costs
| Scenario | Base | Complexity | Total |
|---|---|---|---|
| Simple question, Haiku | 0.5 | 1x | ~0.5 credits |
| Analysis question, Sonnet | 2 | 1.5x | ~3 credits |
| Deep research, Opus | 10 | 2.5x | ~25 credits |
Start with a cheaper model for exploration. Switch to a more capable model for final, important questions.
Model Capabilities
Not all models support all features:
| Capability | Description | Models |
|---|---|---|
| Chat | Basic conversation | All |
| Code Generation | Writing and debugging code | Most |
| Function Calling | Tool use and structured output | GPT-4+, Claude 3+ |
| Vision | Understanding images | GPT-4o, Claude 3+, Gemini |
Tips for Model Selection
Start Cheap, Scale Up
- Try your question with a fast, cheap model (Haiku, Llama 8B)
- If the response isn’t good enough, try a mid-tier model
- Reserve expensive models for truly complex tasks
Match Model to Task
| Task Type | Good Choice | Why |
|---|---|---|
| Summarizing | Mid-tier (Sonnet, GPT-4o) | Needs understanding, not deep reasoning |
| Coding | Sonnet, GPT-4o | Strong at code |
| Complex reasoning | Opus, o3/o4 | Designed for hard problems |
| Quick lookups | Haiku, Llama 8B | Speed over depth |
| Long documents | Gemini Pro | Large context window |
Watch Your Credits
- Check credit costs in the message metadata (info icon)
- Monitor your balance in account settings
- Set up alerts before running low