Llama Models Overview: Unlock AI Potential

info

For detailed information about using models with APIpie, check out our Models Overview and Completions Guide.

Description

The Llama Series represents Meta's family of state-of-the-art large language models. These open-source models, developed by Meta AI, leverage cutting-edge technology to deliver exceptional performance in natural language processing, instruction-following tasks, and extended-context interactions. The models are available through various providers integrated with APIpie's routing system.

Key Features

• Extended Token Capacity: Models support context lengths from 4K to 131K tokens for handling various text processing needs.
• Multi-Provider Availability: Accessible across platforms like OpenRouter, EdenAI, Together, Deepinfra, Monster, and Amazon Bedrock.
• Diverse Applications: Optimized for chat, instruction-following, code generation, vision, and moderation tasks.
• Scalability: Models ranging from efficient 1B parameter configurations to powerful 405B parameter versions.

---

Model List in the Llama Series

Model List updates dynamically please see the Models Route for the up to date list of models

info

For information about model performance and benchmarks, see the Llama Technical Report.

Model Name	Max Tokens	Response Tokens	Provider	Type
nous-hermes-llama2-13b	4,096	4,096	openrouter	llm
llama-2-13b-chat	4,096	4,096	openrouter	llm
llama2-13b-chat-v1	4,096	3,996	bedrock	llm
llama2-70b-chat-v1	4,096	3,996	bedrock	llm
Llama-2-70b-chat-hf	4,096	4,096	deepinfra	llm
CodeLlama-70b-Instruct-hf	4,096	4,096	deepinfra	code
CodeLlama-34b-Instruct-hf	16,384	16,384	deepinfra	code
Phind-CodeLlama-34B-v2	16,384	16,384	deepinfra	code
Llama-2-7b-chat-hf	4,096	4,096	deepinfra	llm
Llama-2-13b-chat-hf	4,096	4,096	deepinfra	llm
Meta-Llama-3-70B-Instruct	8,192	8,192	deepinfra	llm
Meta-Llama-3-8B-Instruct	8,192	8,192	deepinfra	llm
llama-3-lumimaid-8b	24,576	2,048	openrouter	llm
llama-guard-2-8b	8,192	8,192	openrouter	llm
llama-3-lumimaid-70b	8,192	2,048	openrouter	llm
llama3-8b-instruct-v1	8,192	8,092	bedrock	llm
llama3-70b-instruct-v1	8,192	8,092	bedrock	llm
Meta-Llama-3-8B-Instruct	8,192	8,192	monster	llm
llama-3-8b-instruct	8,192	8,192	openrouter	llm
llama-3-70b-instruct	8,192	8,192	openrouter	llm
llama-3.1-sonar-large-128k-online	127,072	127,072	openrouter	llm
llama-3.1-sonar-large-128k-chat	131,072	131,072	openrouter	llm
llama-3.1-sonar-small-128k-online	127,072	127,072	openrouter	llm
llama-3.1-sonar-small-128k-chat	131,072	131,072	openrouter	llm
llama-3.1-sonar-huge-128k-online	127,072	127,072	openrouter	llm
llama-3.1-lumimaid-8b	32,768	2,048	openrouter	llm
meta-llama-3.1-8b-instruct	8,192	8,192	openrouter	llm
llama-3.2-3b-instruct	131,000	131,000	openrouter	llm
llama-3.2-1b-instruct	131,072	131,072	openrouter	llm
llama-3.2-90b-vision-instruct	131,072	131,072	openrouter	vision
llama-3.2-11b-vision-instruct	131,072	4,096	openrouter	vision
llama-3.1-nemotron-70b-instruct	131,000	131,000	openrouter	llm
llama-3.1-lumimaid-70b	16,384	2,048	openrouter	llm
llama-3.3-70b-instruct	131,072	131,072	openrouter	llm
llama3-1-405b-instruct-v1:0	-	-	edenai	llm
llama3-1-70b-instruct-v1:0	-	-	edenai	llm
llama3-1-8b-instruct-v1:0	-	-	edenai	llm
llama3-70b-instruct-v1:0	-	-	edenai	llm
llama3-8b-instruct-v1:0	-	-	edenai	llm
eva-llama-3.33-70b	16,384	4,096	openrouter	llm
Llama-3.1-Nemotron-70B-Instruct-HF	32,768	32,768	together	llm
Llama-3.3-70B-Instruct-Turbo	131,072	131,072	together	llm
Llama-3.2-11B-Vision-Instruct-Turbo	131,072	131,072	together	llm
Llama-3-8b-chat-hf	8,192	8,192	together	llm
Llama-Guard-3-11B-Vision-Turbo	131,072	131,072	together	moderation
Llama-3-70b-chat-hf	8,192	8,192	together	llm
Llama-3.2-3B-Instruct-Turbo	131,072	131,072	together	llm
Meta-Llama-3.1-405B-Instruct-Turbo	130,815	130,815	together	llm
scb10x-llama3-typhoon-v1-5x-4f316	8,192	8,192	together	llm
Meta-Llama-3.1-8B-Instruct-Turbo-128K	131,072	131,072	together	llm
Meta-Llama-3.1-8B-Instruct-Turbo	131,072	131,072	together	llm
Llama-2-13b-chat-hf	4,096	4,096	together	llm
Llama-3.2-90B-Vision-Instruct-Turbo	131,072	131,072	together	llm
Meta-Llama-3.1-70B-Instruct-Turbo	131,072	131,072	together	llm
Meta-Llama-3-8B-Instruct-Turbo	8,192	8,192	together	llm
Meta-Llama-3-8B-Instruct-Lite	8,192	8,192	together	llm
scb10x-llama3-typhoon-v1-5-8b-instruct	8,192	8,192	together	llm
Llama-2-70b-hf	4,096	4,096	together	llm
LlamaGuard-2-8b	8,192	8,192	together	moderation
Llama-Guard-7b	4,096	4,096	together	moderation
Meta-Llama-3-70B-Instruct-Turbo	8,192	8,192	together	llm
Meta-Llama-3-70B-Instruct-Lite	8,192	8,192	together	llm
Llama-Rank-V1	8,192	8,192	together	llm
Llama-2-7b-chat-hf	4,096	4,096	together	llm
Meta-Llama-Guard-3-8B	8,192	8,192	together	moderation
llama3-1-8b-instruct-v1	128,000	128,000	bedrock	llm
llama3-1-70b-instruct-v1	128,000	128,000	bedrock	llm
llama3-2-11b-instruct-v1	128,000	128,000	bedrock	llm
llama3-2-90b-instruct-v1	128,000	128,000	bedrock	llm
llama3-2-1b-instruct-v1	128,000	128,000	bedrock	llm
llama3-2-3b-instruct-v1	8,192	8,092	bedrock	llm
llama3-3-70b-instruct-v1	8,192	8,092	bedrock	llm

---

Example API Call

Below is an example of how to use the Chat Completions API to interact with a model from the Llama Series, such as llama-3-70b-instruct.

curl -L -X POST 'https://apipie.ai/v1/chat/completions' \
-H 'Content-Type: application/json' \
-H 'Accept: application/json' \
-H 'Authorization: Bearer <YOUR_API_KEY>' \
--data-raw '{
  "provider": "openrouter",
  "model": "llama-3-70b-instruct",
  "max_tokens": 150,
  "messages": [
    {
      "role": "user",
      "content": "Can you explain how photosynthesis works?"
    }
  ]
}'

---

Response Example

The expected response structure for the Llama model might look like this:

{
  "id": "chatcmpl-12345example12345",
  "object": "chat.completion",
  "created": 1729535643,
  "provider": "openrouter",
  "model": "llama-3-70b-instruct",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Photosynthesis is the process by which green plants, algae, and some bacteria convert light energy into chemical energy. Here's how it works:\n\n1. **Light Absorption**: Plants capture light energy using a pigment called chlorophyll, which is found in chloroplasts.\n\n2. **Water and Carbon Dioxide**: They absorb water through their roots and carbon dioxide from the air.\n\n3. **Glucose Production**: The light energy is used to convert water and carbon dioxide into glucose (a sugar) and oxygen. The equation is:\n   \n   6CO2 + 6H2O + light energy → C6H12O6 + 6O2\n\nThis process provides energy for the plant and releases oxygen into the atmosphere."
      },
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 125,
    "total_tokens": 140,
    "prompt_characters": 45,
    "response_characters": 520,
    "cost": 0.002250,
    "latency_ms": 3100
  },
  "system_fingerprint": "fp_123abc456def"
}

---

API Highlights

• Provider: Specify the provider or leave blank for automatic selection.
• Model: Use any model from the Llama Series, such as llama-3-70b-instruct or others suited to your task. See Models Guide.
• Max Tokens: Set the maximum response token count (e.g., 150 in this example).
• Messages: Format your request with a sequence of messages, including user input and system instructions. See message formatting.

This example demonstrates how to seamlessly query models from the Llama Series for conversational or instructional tasks.

---

Applications and Integrations

• Conversational AI: Powering chatbots, virtual assistants, and other dialogue-based systems. Try it with LibreChat or OpenWebUI.
• Code Generation: Using CodeLlama variants for programming tasks, code completion, and technical documentation.
• Content Moderation: Leveraging LlamaGuard models for content filtering and safety checks.
• Vision Tasks: Using vision-enabled models for image understanding and multimodal applications.
• Extended Context Tasks: Processing long documents with models supporting up to 131K tokens. Learn more in our Models Guide.

---

Ethical Considerations

The Llama models are powerful tools that should be used responsibly. Users should implement appropriate safeguards and consider potential biases in model outputs. For guidance on responsible AI usage, see Meta's AI Responsibility Guidelines.

---

Licensing

The Llama Series is available under Meta's community license agreement, which allows for both research and commercial use under specific terms. For detailed licensing information, consult the official Llama repository, model cards, and respective hosting providers.

tip

Try out the Llama models in APIpie's various supported integrations.