groq

groq provides ultra-fast inference for open-weight AI models through their specialized hardware architecture. All models are available through AvalAI's unified API with industry-leading performance and competitive pricing.

Available Models

• Safety & Content Moderation

  • groq.llama-guard-4-12b - Advanced content moderation
  • groq.llama-prompt-guard-2-22m - Lightweight prompt injection detection
  • groq.llama-prompt-guard-2-86m - Enhanced prompt injection detection

• Large Language Models

  • groq.llama-4-maverick-17b-128e-instruct - Advanced Llama 4 with 128 experts
  • groq.llama-4-scout-17b-16e-instruct - Efficient Llama 4 with 16 experts
  • groq.kimi-k2-instruct-0905 - Kimi K2 instruction-following
  • groq.gpt-oss-120b - Large open-source GPT (120B)
  • groq.gpt-oss-20b - Efficient open-source GPT (20B)
  • groq.gpt-oss-safeguard-20b - Safety-enhanced GPT (20B)
  • groq.qwen3-32b - Qwen 3 multilingual (32B)

• Audio Models

  • groq.playai-tts - High-quality text-to-speech
  • groq.playai-tts-arabic - Arabic-optimized TTS
  • groq.whisper-large-v3 - Advanced speech recognition
  • groq.whisper-large-v3-turbo - Faster speech recognition

Key Features

• Ultra-Fast Inference: Industry-leading inference speeds powered by groq's custom LPU™ (Language Processing Unit) hardware
• Open-Weight Models: Access to popular open-source models with production-ready performance
• Competitive Pricing: Cost-effective options with prompt caching support for additional savings
• Diverse Capabilities: From safety moderation to multilingual text generation and audio processing
• Production Ready: Enterprise-grade reliability and performance

API Endpoint Support

Model Category	v1/chat/completions	v1/audio/speech	v1/audio/transcriptions
LLMs & Safety	✅ Full	❌	❌
TTS Models	❌	✅ Full	❌
Whisper Models	❌	❌	✅ Full

---

Safety & Content Moderation Models

groq.llama-guard-4-12b

Advanced content moderation model based on Llama Guard 4 with 12 billion parameters.

Pricing: $0.20/1M input tokens, $0.10/1M cached input, $0.20/1M output tokens

Bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "groq.llama-guard-4-12b",
    "messages": [
      {
        "role": "user",
        "content": "Check if this content is safe: [content to moderate]"
      }
    ]
  }'

Python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

response = client.chat.completions.create(
    model="groq.llama-guard-4-12b",
    messages=[
        {
            "role": "user",
            "content": "Check if this content is safe: [content to moderate]",
        }
    ],
)

print(response.choices[0].message.content)

JavaScript

import { OpenAI } from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const response = await client.chat.completions.create({
  model: "groq.llama-guard-4-12b",
  messages: [
    {
      role: "user",
      content: "Check if this content is safe: [content to moderate]",
    },
  ],
});

console.log(response.choices[0].message.content);

Responses API version This version uses `gpt-5.5` because `groq.llama-guard-4-12b` may not be enabled for `/v1/responses` in the current AvalAI model data.

Use this version when the selected model supports /v1/responses. messages moves to input, and the final text is read from response.output_text.

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AVALAI_API_KEY"],
    base_url="https://api.avalai.ir/v1",
)

response = client.responses.create(
    model="gpt-5.5",
    instructions="You are a helpful assistant.",
    input="Check if this content is safe: [content to moderate]",
)

print(response.output_text)

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const response = await client.responses.create({
  model: "gpt-5.5",
  instructions: "You are a helpful assistant.",
  input: "Check if this content is safe: [content to moderate]",
});

console.log(response.output_text);

Bash

curl https://api.avalai.ir/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '
  {
    "model": "gpt-5.5",
    "input": "Check if this content is safe: [content to moderate]",
    "instructions": "You are a helpful assistant."
  }'

• messages → input
• system message → instructions or a developer item
• choices[0].message.content → response.output_text
• for tools and multimodal output, inspect response.output by item type.

groq.llama-prompt-guard-2-22m

Lightweight prompt injection detection model with 22 million parameters.

Pricing: $0.03/1M input tokens, $0.015/1M cached input, $0.03/1M output tokens

groq.llama-prompt-guard-2-86m

Enhanced prompt injection detection model with 86 million parameters.

Pricing: $0.04/1M input tokens, $0.02/1M cached input, $0.04/1M output tokens

---

Large Language Models

groq.llama-4-maverick-17b-128e-instruct

Advanced Llama 4 model with 17 billion parameters and 128 experts, optimized for complex reasoning and instruction following.

Pricing: $0.20/1M input tokens, $0.10/1M cached input, $0.60/1M output tokens

Bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "groq.llama-4-maverick-17b-128e-instruct",
    "messages": [
      {
        "role": "user",
        "content": "Write a Python function to calculate prime numbers."
      }
    ]
  }'

Python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

response = client.chat.completions.create(
    model="groq.llama-4-maverick-17b-128e-instruct",
    messages=[
        {
            "role": "user",
            "content": "Write a Python function to calculate prime numbers.",
        }
    ],
)

print(response.choices[0].message.content)

JavaScript

import { OpenAI } from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const response = await client.chat.completions.create({
  model: "groq.llama-4-maverick-17b-128e-instruct",
  messages: [
    {
      role: "user",
      content: "Write a Python function to calculate prime numbers.",
    },
  ],
});

console.log(response.choices[0].message.content);

Responses API version This version uses `gpt-5.5` because `groq.llama-4-maverick-17b-128e-instruct` may not be enabled for `/v1/responses` in the current AvalAI model data.

Use this version when the selected model supports /v1/responses. messages moves to input, and the final text is read from response.output_text.

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AVALAI_API_KEY"],
    base_url="https://api.avalai.ir/v1",
)

response = client.responses.create(
    model="gpt-5.5",
    instructions="You are a helpful assistant.",
    input="Write a Python function to calculate prime numbers.",
)

print(response.output_text)

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const response = await client.responses.create({
  model: "gpt-5.5",
  instructions: "You are a helpful assistant.",
  input: "Write a Python function to calculate prime numbers.",
});

console.log(response.output_text);

Bash

curl https://api.avalai.ir/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '
  {
    "model": "gpt-5.5",
    "input": "Write a Python function to calculate prime numbers.",
    "instructions": "You are a helpful assistant."
  }'

• messages → input
• system message → instructions or a developer item
• choices[0].message.content → response.output_text
• for tools and multimodal output, inspect response.output by item type.

groq.llama-4-scout-17b-16e-instruct

Efficient Llama 4 model with 17 billion parameters and 16 experts.

Pricing: $0.11/1M input tokens, $0.055/1M cached input, $0.34/1M output tokens

groq.kimi-k2-instruct-0905

Kimi K2 instruction-following model optimized for following complex instructions.

Pricing: $1.00/1M input tokens, $0.50/1M cached input, $0.34/1M output tokens

groq.gpt-oss-120b

Large open-source GPT model with 120 billion parameters.

Pricing: $0.15/1M input tokens, $0.075/1M cached input, $0.75/1M output tokens

groq.gpt-oss-20b

Efficient open-source GPT model with 20 billion parameters.

Pricing: $0.075/1M input tokens, $0.0375/1M cached input, $0.30/1M output tokens

groq.gpt-oss-safeguard-20b

Safety-enhanced GPT model with 20 billion parameters and built-in content filtering.

Pricing: $0.075/1M input tokens, $0.0375/1M cached input, $0.30/1M output tokens

groq.qwen3-32b

Qwen 3 multilingual model with 32 billion parameters, optimized for multiple languages.

Pricing: $0.29/1M input tokens, $0.145/1M cached input, $0.59/1M output tokens

---

Audio Models

groq.playai-tts

High-quality text-to-speech synthesis with natural-sounding voices.

Pricing: $0.00005 per character

Supported Voices: Aaliyah-PlayAI, Adelaide-PlayAI, Angelo-PlayAI, Arista-PlayAI, Atlas-PlayAI, Basil-PlayAI, Briggs-PlayAI, Calum-PlayAI, Celeste-PlayAI, Cheyenne-PlayAI, Chip-PlayAI, Cillian-PlayAI, Deedee-PlayAI, Eleanor-PlayAI, Fritz-PlayAI, Gail-PlayAI, Indigo-PlayAI, Jennifer-PlayAI, Judy-PlayAI, Mamaw-PlayAI, Mason-PlayAI, Mikail-PlayAI, Mitch-PlayAI, Nia-PlayAI, Quinn-PlayAI, Ruby-PlayAI, Thunder-PlayAI

Bash

curl https://api.avalai.ir/v1/audio/speech \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "groq.playai-tts",
    "input": "Hello, this is a test of the PlayAI text-to-speech system.",
    "voice": "Aaliyah-PlayAI"
  }'

Python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

response = client.audio.speech.create(
    model="groq.playai-tts",
    voice="Aaliyah-PlayAI",
    input="Hello, this is a test of the PlayAI text-to-speech system.",
)

response.stream_to_file("output.mp3")

JavaScript

import { OpenAI } from "openai";
import fs from "fs";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const mp3 = await client.audio.speech.create({
  model: "groq.playai-tts",
  voice: "Aaliyah-PlayAI",
  input: "Hello, this is a test of the PlayAI text-to-speech system.",
});

const buffer = Buffer.from(await mp3.arrayBuffer());
await fs.promises.writeFile("output.mp3", buffer);

groq.playai-tts-arabic

Arabic-optimized text-to-speech synthesis.

Pricing: $0.00005 per character

Supported Voices: Same as groq.playai-tts

groq.whisper-large-v3

Advanced speech recognition model based on OpenAI's Whisper Large v3.

Pricing: $0.000031 per second of audio ($0.00185 per minute)

Bash

curl https://api.avalai.ir/v1/audio/transcriptions \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -F file="@audio.mp3" \
  -F model="groq.whisper-large-v3"

Python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

with open("audio.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="groq.whisper-large-v3",
        file=audio_file,
    )

print(transcript.text)

JavaScript

import { OpenAI } from "openai";
import fs from "fs";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const transcript = await client.audio.transcriptions.create({
  model: "groq.whisper-large-v3",
  file: fs.createReadStream("audio.mp3"),
});

console.log(transcript.text);

groq.whisper-large-v3-turbo

Faster speech recognition with maintained accuracy.

Pricing: $0.00001111 per second of audio ($0.000067 per minute)

---

Related Resources

• groq Provider Support News
• API Pricing
• Model Selection Guide
• Audio Processing Guide