The system
How the system worksThe DirectoryDispatchThe LedgerAI modelsAI agentsYour installationThe appThe research libraryWhat changes for your businessUse it from your softwareEverything it can do
Messaging
iMessageWhatsApp
Ads and leads
AdsCreativeLeads
Devices
Devices
Industries
RealtorsRestaurantsHome services
Explore
How it worksPermissions and limitsCompareSet upQuestionsText me

Free model inventory across 7 platforms + CommandCode: 4 key-gated, 2 keyless, 15 OpenRouter free, 3 recommended coders

Six platform model lists fetched 2026-10-08. Four require authentication (Groq 401, Cerebras 403, Together 401, Mistral 401). Two return keyless (Hugging Face 200 with 268 entries, NVIDIA NIM 200 with 80 IDs). OpenRouter has 15 :free variants. CommandCode: free tier exists, OpenAI-compatible API, cline -P commandcode works.

Platform results

PlatformModel list statusFree models verifiedAPI shapecline support
OpenRouter20015 :free variants (see table below)POST https://openrouter.ai/api/v1/chat/completionsYes: cline -P openrouter
Groq401 Invalid API KeyNone — inventory blockedPOST https://api.groq.com/openai/v1/chat/completionsYes: cline -P groq
Cerebras403 Not authenticatedNone — inventory blockedPOST https://api.cerebras.ai/v1/chat/completionsYes: cline -P cerebras
Together401 Missing API keyNone — inventory blockedPOST https://api.together.xyz/v1/chat/completionsYes: cline -P together
Hugging Face200, 268 entries (filter=inference)No free hosted coders — Hub tag list onlyPOST https://router.huggingface.co/v1/chat/completionsYes: cline -P huggingface
Mistral401 Invalid API KeyNone — inventory blockedPOST https://api.mistral.ai/v1/chat/completionsYes: cline -P mistral
NVIDIA NIM200, 80 IDsFree trial credits per FAQ; ongoing free not establishedPOST https://integrate.api.nvidia.com/v1/chat/completionsYes: cline -P nvidia

OpenRouter free models (15)

All fetched 2026-10-08 from /api/v1/models, flagged starred: true in openrouter-free-models-2026-10-08.json:

  • dots-studio/dots-3-note-preview:free — brain/default
  • cohere/north-mini-code:free — coder/default
  • qwen/qwen-2.5-coder-32b-instruct:free
  • qwen/qwen-2.5-72b-instruct:free
  • meta-llama/llama-3.2-3b-instruct:free
  • meta-llama/llama-3.2-1b-instruct:free
  • google/gemma-2-9b-it:free
  • google/gemma-2-2b-it:free
  • microsoft/phi-3-mini-128k-instruct:free
  • microsoft/phi-3-medium-128k-instruct:free
  • nvidia/nemotron-3-ultra:free
  • nvidia/nemotron-3-ultra-550b-a55b:free
  • mistralai/mistral-7b-instruct:free
  • mistralai/mistral-nemo:free
  • gryphe/mythomax-l2-13b:free

CommandCode evaluation

Source: /Users/cyrusmassoumi/misc-sheets/src/commandcode-chat.js

  • Free tier: yes, no credit card required
  • API: OpenAI-compatible chat completions at https://api.commandcode.com/v1/chat/completions
  • Auth: Bearer token (API key from dashboard)
  • Models: commandcode-default, commandcode-reasoning, commandcode-fast
  • cline support: cline -P commandcode -m <model> (provider installed)
  • Rate limits: not published in source; dashboard shows per-model limits
  • node --check src/commandcode-chat.js: passes

Best 3 free coders (ranked)

  1. cohere/north-mini-code:free (OpenRouter) — designated coder default, 128k context, zero token price
  2. qwen/qwen-2.5-coder-32b-instruct:free (OpenRouter) — 32B coding specialist, 128k context
  3. commandcode-default (CommandCode) — native free tier, no card, OpenAI-compatible, cline-ready

Evidence sources (all fetched 2026-10-08)

OpenRouter: https://openrouter.ai/api/v1/models Groq: https://api.groq.com/openai/v1/models Cerebras: https://api.cerebras.ai/v1/models Together: https://api.together.xyz/v1/models Hugging Face: https://huggingface.co/api/models?filter=inference Mistral: https://api.mistral.ai/v1/models NVIDIA NIM: https://integrate.api.nvidia.com/v1/models CommandCode source: src/commandcode-chat.js OpenRouter free models file: data/openrouter-free-models-2026-10-08.json