Language Models Overview¶
This page provides a comprehensive overview of all supported language models in the unique_toolkit.
Model Properties¶
The following properties are documented for each model:
- Name: The display name of the model
- Provider: The service provider (Azure, LiteLLM, etc.)
- Version: Model version information
- Encoder: Tokenizer/encoder used within the code for the model
- Token Limits: Input, output, and total token limits
- Capabilities: Supported features (streaming, function calling, etc.)
- Temperature Bounds: Min/max temperature settings (if available)
Quick Reference¶
Token Limits Summary¶
| Model | Provider | Input Tokens | Output Tokens | Total Tokens |
|---|---|---|---|---|
| AZURE_GPT_35_TURBO_0125 | AZURE | 16,385 | 4,096 | 20,481 |
| AZURE_GPT_41_2025_0414 | AZURE | 1,047,576 | 32,768 | 1,080,344 |
| AZURE_GPT_41_MINI_2025_0414 | AZURE | 1,047,576 | 32,768 | 1,080,344 |
| AZURE_GPT_41_NANO_2025_0414 | AZURE | 1,047,576 | 32,768 | 1,080,344 |
| AZURE_GPT_45_PREVIEW_2025_0227 | AZURE | 128,000 | 16,384 | 144,384 |
| AZURE_GPT_4_0613 | AZURE | 3,276 | 4,915 | 8,191 |
| AZURE_GPT_4_32K_0613 | AZURE | 13,107 | 19,660 | 32,767 |
| AZURE_GPT_4_TURBO_2024_0409 | AZURE | 128,000 | 4,096 | 132,096 |
| AZURE_GPT_4o_2024_0513 | AZURE | 128,000 | 4,096 | 132,096 |
| AZURE_GPT_4o_2024_0806 | AZURE | 128,000 | 16,384 | 144,384 |
| AZURE_GPT_4o_2024_1120 | AZURE | 128,000 | 16,384 | 144,384 |
| AZURE_GPT_4o_MINI_2024_0718 | AZURE | 128,000 | 16,384 | 144,384 |
| AZURE_GPT_51_2025_1113 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_51_CHAT_2025_1113 | AZURE | 128,000 | 16,384 | 144,384 |
| AZURE_GPT_51_CODEX_2025_1113 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_51_CODEX_MINI_2025_1113 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_51_THINKING_2025_1113 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_5_2025_0807 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_5_CHAT_2025_0807 | AZURE | 128,000 | 16,384 | 144,384 |
| AZURE_GPT_5_MINI_2025_0807 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_5_NANO_2025_0807 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_GPT_5_PRO_2025_1006 | AZURE | 272,000 | 128,000 | 400,000 |
| AZURE_o1_2024_1217 | AZURE | 200,000 | 100,000 | 300,000 |
| AZURE_o1_MINI_2024_0912 | AZURE | 128,000 | 65,536 | 193,536 |
| AZURE_o3_2025_0416 | AZURE | 200,000 | 100,000 | 300,000 |
| AZURE_o3_MINI_2025_0131 | AZURE | 200,000 | 100,000 | 300,000 |
| AZURE_o4_MINI_2025_0416 | AZURE | 200,000 | 100,000 | 300,000 |
| litellm:anthropic-claude-3-7-sonnet | LITELLM | 180,000 | 64,000 | 244,000 |
| litellm:anthropic-claude-3-7-sonnet-thinking | LITELLM | 180,000 | 64,000 | 244,000 |
| litellm:anthropic-claude-haiku-4-5 | LITELLM | 180,000 | 64,000 | 244,000 |
| litellm:anthropic-claude-opus-4 | LITELLM | 180,000 | 32,000 | 212,000 |
| litellm:anthropic-claude-opus-4-1 | LITELLM | 180,000 | 32,000 | 212,000 |
| litellm:anthropic-claude-sonnet-4 | LITELLM | 180,000 | 64,000 | 244,000 |
| litellm:anthropic-claude-sonnet-4-5 | LITELLM | 180,000 | 64,000 | 244,000 |
| litellm:deepseek-r1 | LITELLM | 64,000 | 4,000 | 68,000 |
| litellm:deepseek-v3-1 | LITELLM | 128,000 | 4,000 | 132,000 |
| litellm:gemini-2-0-flash | LITELLM | 1,048,576 | 8,192 | 1,056,768 |
| litellm:gemini-2-5-flash | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:gemini-2-5-flash-lite | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:gemini-2-5-flash-lite-preview-06-17 | LITELLM | 1,000,000 | 64,000 | 1,064,000 |
| litellm:gemini-2-5-flash-preview-05-20 | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:gemini-2-5-pro | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:gemini-2-5-pro-exp-03-25 | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:gemini-2-5-pro-preview-06-05 | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:gemini-3-pro-preview | LITELLM | 1,048,576 | 65,536 | 1,114,112 |
| litellm:openai-gpt-4-1-mini | LITELLM | 1,047,576 | 32,768 | 1,080,344 |
| litellm:openai-gpt-4-1-nano | LITELLM | 1,047,576 | 32,768 | 1,080,344 |
| litellm:openai-gpt-5 | LITELLM | 272,000 | 128,000 | 400,000 |
| litellm:openai-gpt-5-1 | LITELLM | 272,000 | 128,000 | 400,000 |
| litellm:openai-gpt-5-1-thinking | LITELLM | 272,000 | 128,000 | 400,000 |
| litellm:openai-gpt-5-chat | LITELLM | 128,000 | 16,384 | 144,384 |
| litellm:openai-gpt-5-mini | LITELLM | 272,000 | 128,000 | 400,000 |
| litellm:openai-gpt-5-nano | LITELLM | 272,000 | 128,000 | 400,000 |
| litellm:openai-gpt-5-pro | LITELLM | 272,000 | 128,000 | 400,000 |
| litellm:openai-o1 | LITELLM | 200,000 | 100,000 | 300,000 |
| litellm:openai-o3 | LITELLM | 200,000 | 100,000 | 300,000 |
| litellm:openai-o3-deep-research | LITELLM | 200,000 | 100,000 | 300,000 |
| litellm:openai-o3-pro | LITELLM | 200,000 | 100,000 | 300,000 |
| litellm:openai-o4-mini | LITELLM | 200,000 | 100,000 | 300,000 |
| litellm:openai-o4-mini-deep-research | LITELLM | 200,000 | 100,000 | 300,000 |
| litellm:qwen-3-235B-A22B | LITELLM | 256,000 | 32,768 | 288,768 |
| litellm:qwen-3-235B-A22B-thinking | LITELLM | 256,000 | 32,768 | 288,768 |
Capabilities Matrix¶
To use any of these models in your application:
Model Selection Guide¶
For High-Volume Applications¶
- Cost-effective: GPT-4o Mini, GPT-5 Mini, Claude 3.7 Sonnet
- Balanced: GPT-4o, GPT-5, Claude Sonnet 4
For Complex Reasoning¶
- Advanced: o1, o3, Claude 3.7 Sonnet Thinking
- Research: o3 Deep Research, o4 Mini Deep Research
For Function Calling¶
- Reliable: GPT-4o, GPT-5, Claude Sonnet 4
- Fast: GPT-4o Mini, GPT-5 Mini
For Structured Output¶
- All modern models support structured output capabilities
Last updated: 2025-11-24