llama-3.1-8b-instruct-fast
Text Generation • Meta[Fast version] The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models. The Llama 3.1 instruction tuned text only models are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.
| Features | |
|---|---|
| Context Window ↗ | 128,000 tokens |
| Terms and License | link ↗ |
Playground
Try out this model with Workers AI LLM Playground. It does not require any setup or authentication and an instant way to preview and test a model directly in the browser.
Launch the LLM PlaygroundUsage
Worker - Streaming
export interface Env { AI: Ai;}
export default { async fetch(request, env): Promise<Response> {
const messages = [ { role: "system", content: "You are a friendly assistant" }, { role: "user", content: "What is the origin of the phrase Hello, World", }, ];
const stream = await env.AI.run("@cf/meta/llama-3.1-8b-instruct-fast", { messages, stream: true, });
return new Response(stream, { headers: { "content-type": "text/event-stream" }, }); },} satisfies ExportedHandler<Env>;Worker
export interface Env { AI: Ai;}
export default { async fetch(request, env): Promise<Response> {
const messages = [ { role: "system", content: "You are a friendly assistant" }, { role: "user", content: "What is the origin of the phrase Hello, World", }, ]; const response = await env.AI.run("@cf/meta/llama-3.1-8b-instruct-fast", { messages });
return Response.json(response); },} satisfies ExportedHandler<Env>;Python
import osimport requests
ACCOUNT_ID = "your-account-id"AUTH_TOKEN = os.environ.get("CLOUDFLARE_AUTH_TOKEN")
prompt = "Tell me all about PEP-8"response = requests.post( f"https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/run/@cf/meta/llama-3.1-8b-instruct-fast", headers={"Authorization": f"Bearer {AUTH_TOKEN}"}, json={ "messages": [ {"role": "system", "content": "You are a friendly assistant"}, {"role": "user", "content": prompt} ] })result = response.json()print(result)curl
curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run/@cf/meta/llama-3.1-8b-instruct-fast \ -X POST \ -H "Authorization: Bearer $CLOUDFLARE_AUTH_TOKEN" \ -d '{ "messages": [{ "role": "system", "content": "You are a friendly assistant" }, { "role": "user", "content": "Why is pizza so good" }]}'Parameters
* indicates a required field
Input
- Prompt object
-
frequency_penaltynumber min 0 max 2Decreases the likelihood of the model repeating the same lines verbatim.
-
imageone of-
0arrayAn array of integers that represent the image data constrained to 8-bit unsigned integer values
-
itemsnumberA value between 0 and 255
-
-
1stringBinary string representing the image contents.
-
-
lorastringName of the LoRA (Low-Rank Adaptation) model to fine-tune the base model.
-
max_tokensinteger default 256The maximum number of tokens to generate in the response.
-
presence_penaltynumber min 0 max 2Increases the likelihood of the model introducing new topics.
-
prompt *string min 1 max 131072The input text prompt for the model to generate a response.
-
rawbooleanIf true, a chat template is not applied and you must adhere to the specific model's expected formatting.
-
repetition_penaltynumber min 0 max 2Penalty for repeated tokens; higher values discourage repetition.
-
seedinteger min 1 max 9999999999Random seed for reproducibility of the generation.
-
streambooleanIf true, the response will be streamed back incrementally using SSE, Server Sent Events.
-
temperaturenumber default 0.6 min 0 max 5Controls the randomness of the output; higher values produce more random results.
-
top_kinteger min 1 max 50Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.
-
top_pnumber min 0 max 2Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.
-
- Messages object
-
frequency_penaltynumber min 0 max 2Decreases the likelihood of the model repeating the same lines verbatim.
-
functionsarray-
itemsobject-
code *string -
name *string
-
-
-
imageone of-
0arrayAn array of integers that represent the image data constrained to 8-bit unsigned integer values
-
itemsnumberA value between 0 and 255
-
-
1stringBinary string representing the image contents.
-
-
max_tokensinteger default 256The maximum number of tokens to generate in the response.
-
messages *arrayAn array of message objects representing the conversation history.
-
itemsobject-
content *string max 131072The content of the message as a string.
-
role *stringThe role of the message sender (e.g., 'user', 'assistant', 'system', 'tool').
-
-
-
presence_penaltynumber min 0 max 2Increases the likelihood of the model introducing new topics.
-
repetition_penaltynumber min 0 max 2Penalty for repeated tokens; higher values discourage repetition.
-
seedinteger min 1 max 9999999999Random seed for reproducibility of the generation.
-
streambooleanIf true, the response will be streamed back incrementally.
-
temperaturenumber default 0.6 min 0 max 5Controls the randomness of the output; higher values produce more random results.
-
toolsarrayA list of tools available for the assistant to use.
-
itemsone of-
0object-
description *stringA brief description of what the tool does.
-
name *stringThe name of the tool. More descriptive the better.
-
parameters *objectSchema defining the parameters accepted by the tool.
-
properties *objectDefinitions of each parameter.
-
additionalPropertiesobject-
description *stringA description of the expected parameter.
-
type *stringThe data type of the parameter.
-
-
-
requiredarrayList of required parameter names.
-
itemsstring
-
-
type *stringThe type of the parameters object (usually 'object').
-
-
-
1object-
function *objectDetails of the function tool.
-
description *stringA brief description of what the function does.
-
name *stringThe name of the function.
-
parameters *objectSchema defining the parameters accepted by the function.
-
properties *objectDefinitions of each parameter.
-
additionalPropertiesobject-
description *stringA description of the expected parameter.
-
type *stringThe data type of the parameter.
-
-
-
requiredarrayList of required parameter names.
-
itemsstring
-
-
type *stringThe type of the parameters object (usually 'object').
-
-
-
type *stringSpecifies the type of tool (e.g., 'function').
-
-
-
-
top_kinteger min 1 max 50Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.
-
top_pnumber min 0 max 2Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.
-
Output
-
0object-
responsestringThe generated text response from the model
-
tool_callsarrayAn array of tool calls requests made during the response generation
-
itemsobject-
argumentsobjectThe arguments passed to be passed to the tool call request
-
namestringThe name of the tool to be called
-
-
-
-
1string
API Schemas
The following schemas are based on JSON Schema
{ "oneOf": [ { "properties": { "frequency_penalty": { "description": "Decreases the likelihood of the model repeating the same lines verbatim.", "maximum": 2, "minimum": 0, "type": "number" }, "image": { "oneOf": [ { "description": "An array of integers that represent the image data constrained to 8-bit unsigned integer values", "items": { "description": "A value between 0 and 255", "type": "number" }, "type": "array" }, { "description": "Binary string representing the image contents.", "format": "binary", "type": "string" } ] }, "lora": { "description": "Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model.", "type": "string" }, "max_tokens": { "default": 256, "description": "The maximum number of tokens to generate in the response.", "type": "integer" }, "presence_penalty": { "description": "Increases the likelihood of the model introducing new topics.", "maximum": 2, "minimum": 0, "type": "number" }, "prompt": { "description": "The input text prompt for the model to generate a response.", "maxLength": 131072, "minLength": 1, "type": "string" }, "raw": { "default": false, "description": "If true, a chat template is not applied and you must adhere to the specific model's expected formatting.", "type": "boolean" }, "repetition_penalty": { "description": "Penalty for repeated tokens; higher values discourage repetition.", "maximum": 2, "minimum": 0, "type": "number" }, "seed": { "description": "Random seed for reproducibility of the generation.", "maximum": 9999999999, "minimum": 1, "type": "integer" }, "stream": { "default": false, "description": "If true, the response will be streamed back incrementally using SSE, Server Sent Events.", "type": "boolean" }, "temperature": { "default": 0.6, "description": "Controls the randomness of the output; higher values produce more random results.", "maximum": 5, "minimum": 0, "type": "number" }, "top_k": { "description": "Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.", "maximum": 50, "minimum": 1, "type": "integer" }, "top_p": { "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.", "maximum": 2, "minimum": 0, "type": "number" } }, "required": [ "prompt" ], "title": "Prompt" }, { "properties": { "frequency_penalty": { "description": "Decreases the likelihood of the model repeating the same lines verbatim.", "maximum": 2, "minimum": 0, "type": "number" }, "functions": { "items": { "properties": { "code": { "type": "string" }, "name": { "type": "string" } }, "required": [ "name", "code" ], "type": "object" }, "type": "array" }, "image": { "oneOf": [ { "description": "An array of integers that represent the image data constrained to 8-bit unsigned integer values", "items": { "description": "A value between 0 and 255", "type": "number" }, "type": "array" }, { "description": "Binary string representing the image contents.", "format": "binary", "type": "string" } ] }, "max_tokens": { "default": 256, "description": "The maximum number of tokens to generate in the response.", "type": "integer" }, "messages": { "description": "An array of message objects representing the conversation history.", "items": { "properties": { "content": { "description": "The content of the message as a string.", "maxLength": 131072, "type": "string" }, "role": { "description": "The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool').", "type": "string" } }, "required": [ "role", "content" ], "type": "object" }, "type": "array" }, "presence_penalty": { "description": "Increases the likelihood of the model introducing new topics.", "maximum": 2, "minimum": 0, "type": "number" }, "repetition_penalty": { "description": "Penalty for repeated tokens; higher values discourage repetition.", "maximum": 2, "minimum": 0, "type": "number" }, "seed": { "description": "Random seed for reproducibility of the generation.", "maximum": 9999999999, "minimum": 1, "type": "integer" }, "stream": { "default": false, "description": "If true, the response will be streamed back incrementally.", "type": "boolean" }, "temperature": { "default": 0.6, "description": "Controls the randomness of the output; higher values produce more random results.", "maximum": 5, "minimum": 0, "type": "number" }, "tools": { "description": "A list of tools available for the assistant to use.", "items": { "oneOf": [ { "properties": { "description": { "description": "A brief description of what the tool does.", "type": "string" }, "name": { "description": "The name of the tool. More descriptive the better.", "type": "string" }, "parameters": { "description": "Schema defining the parameters accepted by the tool.", "properties": { "properties": { "additionalProperties": { "properties": { "description": { "description": "A description of the expected parameter.", "type": "string" }, "type": { "description": "The data type of the parameter.", "type": "string" } }, "required": [ "type", "description" ], "type": "object" }, "description": "Definitions of each parameter.", "type": "object" }, "required": { "description": "List of required parameter names.", "items": { "type": "string" }, "type": "array" }, "type": { "description": "The type of the parameters object (usually 'object').", "type": "string" } }, "required": [ "type", "properties" ], "type": "object" } }, "required": [ "name", "description", "parameters" ] }, { "properties": { "function": { "description": "Details of the function tool.", "properties": { "description": { "description": "A brief description of what the function does.", "type": "string" }, "name": { "description": "The name of the function.", "type": "string" }, "parameters": { "description": "Schema defining the parameters accepted by the function.", "properties": { "properties": { "additionalProperties": { "properties": { "description": { "description": "A description of the expected parameter.", "type": "string" }, "type": { "description": "The data type of the parameter.", "type": "string" } }, "required": [ "type", "description" ], "type": "object" }, "description": "Definitions of each parameter.", "type": "object" }, "required": { "description": "List of required parameter names.", "items": { "type": "string" }, "type": "array" }, "type": { "description": "The type of the parameters object (usually 'object').", "type": "string" } }, "required": [ "type", "properties" ], "type": "object" } }, "required": [ "name", "description", "parameters" ], "type": "object" }, "type": { "description": "Specifies the type of tool (e.g., 'function').", "type": "string" } }, "required": [ "type", "function" ] } ], "type": "object" }, "type": "array" }, "top_k": { "description": "Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises.", "maximum": 50, "minimum": 1, "type": "integer" }, "top_p": { "description": "Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses.", "maximum": 2, "minimum": 0, "type": "number" } }, "required": [ "messages" ], "title": "Messages" } ], "type": "object"}{ "oneOf": [ { "contentType": "application/json", "properties": { "response": { "description": "The generated text response from the model", "type": "string" }, "tool_calls": { "description": "An array of tool calls requests made during the response generation", "items": { "properties": { "arguments": { "description": "The arguments passed to be passed to the tool call request", "type": "object" }, "name": { "description": "The name of the tool to be called", "type": "string" } }, "type": "object" }, "type": "array" } }, "type": "object" }, { "contentType": "text/event-stream", "format": "binary", "type": "string" } ]}