freeride-h

name	freeride-h
description	Multi-provider free-model manager for Hermes. Browse, switch, and auto-rotate across 9 providers and 43+ free AI models via the freeride CLI. Triggers: user asks about free models, wants to switch models mid-task, mentions big-pickle/qwen3.5/llama/local, watcher daemon, quota exhaustion, or model rotation.

FreeRide-H — Free Model Manager

PyPI: pip install freeride-h Source: pip install git+https://github.com/dazeb/freeride-h.git Docs: freeride list | freeride --help

Browse, switch, and auto-rotate across 9 providers and 43+ free AI models.

Quick Start

# Install
pip install freeride-h

# Browse free models
freeride list                        # all 46 models
freeride free                        # only free-tier (43 models)
freeride list --provider ollama      # filter by provider

# Switch models mid-task
freeride switch big-pickle           # OpenCode Zen
freeride switch qwen3.5:9b           # Ollama Local (GPU, no API key)
freeride switch codestral            # Mistral (256K ctx)
freeride switch llama-3.3-70b        # Groq (fast inference)
freeride switch nemotron             # NVIDIA NIM / OpenRouter

# Configure all free providers
freeride auto

# Background auto-rotation daemon
freeride-watcher                     # probes every 60s, rotates on fail/slow
freeride-watcher --once              # single check-and-rotate
freeride-watcher --status            # show rotation state

Provider Catalog

For the full provider catalog with all model IDs, context lengths, and API endpoints, see references/provider-catalog.md.

Cloud Free Tier (need API key)

Provider	Env Var	Free Tier	Models
OpenRouter	OPENROUTER_API_KEY	~6M/mo per model	owl-alpha, elephant-alpha, hy3-preview, nemotron-3-120b, ring-2.6-1t, pareto-code
OpenCode Zen	OPENCODE_ZEN_API_KEY	Free tier	big-pickle, deepseek-v4-flash-free, qwen3.6-plus-free, minimax-m2.5-free, nemotron-3-super-free
Mistral	MISTRAL_API_KEY	~1B/mo shared	Mistral Large, Ministral 8B, Codestral (256K ctx), Mistral Saba
Groq	GROQ_API_KEY	~30M/mo per model	Llama 3.3 70B, Llama 4 Scout, Qwen3 32B, GPT-OSS 120B
Cerebras	CEREBRAS_API_KEY	~30M/mo shared	Qwen3 235B, GPT-OSS 120B, Llama 3.1 8B
SambaNova	SAMBANOVA_API_KEY	~3M/mo shared	DeepSeek V3.1/V3.2, Llama 4 Maverick, Llama 3.3 70B, Gemma 3 12B, GPT-OSS 120B
NVIDIA NIM	NVIDIA_API_KEY	Credit-based	Llama 3.1 70B, Nemotron 4 340B, Mistral Large

Local (no API key)

Provider	Base URL	Models
Ollama Local	http://localhost:11434/v1	qwen3.5:9b, gemma4-abliterated, nemotron-3-nano, llama3.2:1b (+ more)

Cloud GPU

Provider	Env Var	Models
Ollama Cloud	OLLAMA_API_KEY	GPT-OSS, GLM-4.7, Qwen3, Kimi K2

When to Use

• Model stuck on quota — freeride list → freeride switch <model>
• User asks for a specific model — auto-detects provider from registry
• User wants local inference — freeride switch qwen3.5:9b
• Rate limits — start freeride-watcher as background daemon
• What's available? — freeride free shows all free models with context sizes

Architecture

PROVIDERS = [
  Provider(name="Groq", env_var="GROQ_API_KEY", base_url="...", hermes_name="groq", models=[...]),
  ...
]

To add a provider, append to the PROVIDERS list in freeride.py. The CLI, auto-config, and watcher auto-discover it.

Files

File	Purpose
freeride.py	CLI — list, free, switch, auto, status
watcher.py	Background daemon — probes and auto-rotates
setup.py / pyproject.toml	Pip packaging
__init__.py	Module imports
references/provider-catalog.md	Full model catalog