Available Models
The AI Commons API provides access to multiple state-of-the-art language models. Choose the model that best fits your use case based on complexity, speed, and cost requirements.
Claude Models (Anthropic)
Claude models are known for their strong reasoning capabilities, nuanced understanding, and safety features.
| Model ID | Description | Context | Best For |
|---|---|---|---|
claude-v4.6-opus |
Most capable Claude 4 model | 200K | Complex reasoning, deep analysis |
claude-v4.6-sonnet |
Balanced Claude 4 model | 200K | General purpose, coding, analysis |
claude-v4.5-haiku |
Fast, efficient Claude 4 | 200K | Quick responses, simple tasks |
Amazon Nova Models
Amazon Nova models offer cost-effective solutions with strong performance across various tasks.
| Model ID | Description | Best For |
|---|---|---|
amazon-nova-pro |
High capability Nova | Complex tasks, reasoning |
amazon-nova-lite |
Balanced Nova | General purpose, cost-effective |
amazon-nova-micro |
Fast, efficient Nova | Simple queries, high throughput |
Llama Models (Meta)
Open-source Llama models offering transparency and strong performance.
| Model ID | Description | Best For |
|---|---|---|
llama-4-maverick-17b-instruct |
Llama 4 balanced | General purpose |
llama-4-scout-17b-instruct |
Llama 4 efficient | Quick tasks, efficiency |
Other Models
| Model ID | Description | Best For |
|---|---|---|
qwen3-32b |
Qwen 3 model | Multilingual, general purpose |
OpenAI Models
Open-weight GPT models with built-in safety and content moderation capabilities.
| Model ID | Description | Best For |
|---|---|---|
openai-gpt-oss-20b |
OpenAI GPT-OSS Safeguard 20B | Content moderation, lightweight safety tasks |
openai-gpt-oss-120b |
OpenAI GPT-OSS Safeguard 120B | Advanced safety, general purpose with guardrails |
Choosing the Right Model
For Production Use:
- Best Quality:
claude-v4.6-opus,claude-v4.6-sonnet - Best Speed:
claude-v4.5-haiku,amazon-nova-micro - Best Cost/Performance:
amazon-nova-lite,amazon-nova-micro
By Use Case:
Complex Reasoning
Use claude-v4.6-opus or claude-v4.6-sonnet for tasks requiring deep analysis and multi-step reasoning.
General Purpose
Use claude-v4.6-sonnet or amazon-nova-pro for balanced performance across various tasks.
Quick Responses
Use claude-v4.5-haiku or amazon-nova-micro for fast, simple queries.
Multilingual
Use qwen3-32b for strong multilingual support.
Reasoning Mode
Some models support enhanced reasoning capabilities when enable_reasoning: true is set:
Supported Models:
claude-v4.6-opusclaude-v4.6-sonnet
Example Request:
{
"message": {
"content": [{
"content_type": "text",
"body": "Solve this complex problem step by step"
}],
"model": "claude-v4.6-opus"
},
"enable_reasoning": true
}
Response with Reasoning:
{
"content": [
{
"content_type": "reasoning",
"text": "Let me think through this step by step...",
"signature": "reasoning_signature"
},
{
"content_type": "text",
"body": "Based on my reasoning, the answer is..."
}
]
}
Model Selection Example
import requests
API_URL = "<your-api-url>"
API_KEY = "<your-api-key>"
# For complex analysis tasks
def analyze_with_opus(text):
return create_conversation(
message=text,
model="claude-v4.6-opus",
temperature=0.3 # Lower for analytical tasks
)
# For quick customer service responses
def quick_response(text):
return create_conversation(
message=text,
model="claude-v4.5-haiku",
temperature=0.7,
max_tokens=500
)
# For creative writing
def creative_writing(prompt):
return create_conversation(
message=prompt,
model="claude-v4.6-sonnet",
temperature=0.9 # Higher for creativity
)