Baltor Get started

Endpoints

Groq

A hosted service with its own key and prices. This page lists the addresses it answers, the facts its documentation states, and the setup of each harness.

Addresses and facts

Authentication
Authorization: Bearer, with the key in GROQ_API_KEY provider documentation, read
Structured output
JSON schema output with strict mode on the models the page lists. provider documentation, read
Tool calling
provider documentation, read
Rate limits
Read the page provider documentation, read
Prices
Unknown
Data retention
Inference requests are not retained by default. Batch files are kept for 30 days unless deleted earlier. Read the page provider documentation, read
Documentation
Unknown
Output limits Baltor recorded
Unknown

Harness setup

Replace allam-2-7b with the model you want.

OpenCode

Put this in opencode.json in your project folder:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "groq": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Groq",
      "options": {
        "baseURL": "https://api.groq.com/openai/v1",
        "apiKey": "{env:GROQ_API_KEY}"
      },
      "models": {
        "allam-2-7b": {
          "name": "allam-2-7b"
        }
      }
    }
  }
}
  • OpenCode reads any OpenAI-compatible address through the @ai-sdk/openai-compatible package, and an address that speaks the Responses API through @ai-sdk/openai.

From OpenCode documentation, read .

Pi

Put this in ~/.pi/agent/models.json:

{
  "providers": {
    "groq": {
      "baseUrl": "https://api.groq.com/openai/v1",
      "api": "openai-completions",
      "apiKey": "$GROQ_API_KEY",
      "models": [
        {
          "id": "allam-2-7b"
        }
      ]
    }
  }
}
  • The apiKey field can name an environment variable as $NAME.

From Pi documentation, read .

Codex

Put this in ~/.codex/config.toml:

model = "allam-2-7b"
model_provider = "groq"

[model_providers.groq]
name = "Groq"
base_url = "https://api.groq.com/openai/v1"
env_key = "GROQ_API_KEY"
  • Codex speaks the Responses API only: responses is the one supported wire API of a custom provider. Ollama and LM Studio are built in and start with --oss.

From Codex documentation, read .

Claude Code

Claude Code sends Anthropic Messages requests, and Groq documents no such address. A gateway that translates to that API can sit in between.

  • Claude Code sends Anthropic Messages requests to ANTHROPIC_BASE_URL. Anthropic says it does not support routing Claude Code to models other than Claude through any gateway, so some features may not work with another model.

From Claude Code documentation, read .

Models it lists

Prices in US dollars per million tokens, input and output, as models.dev, read records them.

ModelInputOutputContextAs of
ALLaM-2-7b allam-2-7b0 USD0 USD4,096 older than 30 days
Canopy Labs Orpheus Arabic Saudi canopylabs/orpheus-arabic-saudiUnknownUnknown4,000
Canopy Labs Orpheus V1 English canopylabs/orpheus-v1-englishUnknownUnknown4,000
Compound groq/compoundUnknownUnknown131,072
Compound Mini groq/compound-miniUnknownUnknown131,072
Llama 3.1 8B llama-3.1-8b-instant0.050 USD0.080 USD131,072 older than 30 days
Llama 3.3 70B llama-3.3-70b-versatile0.590 USD0.790 USD131,072 older than 30 days
Llama Prompt Guard 2 22M meta-llama/llama-prompt-guard-2-22m0.030 USD0.030 USD512 older than 30 days
Prompt Guard 2 86M meta-llama/llama-prompt-guard-2-86m0.040 USD0.040 USD512 older than 30 days
GPT OSS 120B openai/gpt-oss-120b0.150 USD0.600 USD131,072 older than 30 days
GPT OSS 20B openai/gpt-oss-20b0.075 USD0.300 USD131,072 older than 30 days
Safety GPT OSS 20B openai/gpt-oss-safeguard-20b0.075 USD0.300 USD131,072 older than 30 days
Qwen3.6 27B qwen/qwen3.6-27b0.600 USD3.00 USD131,072 older than 30 days
Qwen3.8 27B qwen/qwen3.8-27b0.800 USD4.00 USD131,042 older than 30 days
Whisper whisper-large-v3UnknownUnknownUnknown
Whisper Large V3 Turbo whisper-large-v3-turboUnknownUnknownUnknown