VS Code Coding Agents · Cline · Roo Code · Kilo Code · LiteLLM

Cline & Roo Code Free Models: Base URLs, Context Limits & Setup (2026)

VS Code agents like Cline, Roo Code, and Kilo Code send large system prompts and file trees on every turn—which quickly triggers 429s on low-TPM free tiers. Pick any free provider below to inspect context limits, training privacy, and copy exact provider settings.

Choose Provider & Privacy Model ID & Context Window Rate Limits & Training Rule Agent Fit (Cline / Roo / Kilo)
Trains on Free Tier
gemini-3.8-flash
Context: 1M tokens
Free tier (No Card · per-project RPM/TPM/RPD in AI Studio)
Free tier may train models (avoid for proprietary code)
Large repo context in Cline / Roo Code without hitting token truncation
No Training
openai/gpt-oss-120b
Context: 128K tokens (8K TPM free cap)
30 RPM · 1,000 RPD · 8K TPM · 200K TPD (No Card)
No training on API prompts by default
Fast Ask/Architect turns & small diffs (keep system prompts compact under 8K TPM)
Trains on Free Tier
google/gemma-4-31b-it:free
Context: 128K+ tokens (varies by model)
20 RPM · 50 RPD (<10 credits) or 1,000 RPD (>=10 credits)
Varies by upstream :free provider (often logs prompts)
1-click native OpenRouter integration inside Cline, Roo Code & Kilo Code
Trains on Free Tier
codestral-latest
Context: 256K tokens
1 RPS (~60 RPM) · Experiment Plan (Phone SMS, No Card)
Prompts opted into training on Free Experiment tier
Code refactoring & autocomplete with strict 1 req/sec pacing
No Training
meta/llama-3.3-70b-instruct
Context: 128K tokens
1,000 Starter Credits · ~40 RPM cap (Phone SMS, No Card)
No training on API payloads
Private prototype sprints with 40 RPM burst speed
No Training
gpt-oss-120b
Context: 128K tokens
20 RPM · 20 RPD · 200K TPD (No Card)
Zero data retention on inference API
Zero-data-retention secondary fallback profile in Roo Code / LiteLLM

Interactive Settings Configurator (Cline / Roo Code / Kilo Code / LiteLLM)

API Provider Setting
Google Gemini (or OpenAI Compatible)
OpenAI-Compatible Base URL
https://generativelanguage.googleapis.com/v1beta/openai/
Model ID
gemini-3.8-flash
Context & Privacy Watchout
1M tokens · Free tier may train models

      

⚠️ Three Gotchas When Running Cline / Roo Code on Free APIs

  • TPM Truncation (Groq 8K TPM): Cline's built-in tool instructions alone are several thousand tokens. Set a compact custom instruction or route large context turns to Google AI Studio (1M context).
  • Commercial Code Privacy: Google AI Studio Free, Mistral Experiment, and many OpenRouter :free providers may retain prompts for training. Use NVIDIA NIM, SambaNova, or paid zero-retention endpoints for client repos.
  • OpenRouter 50 vs 1,000 RPD: One coding task can fire 15–30 agent steps. If you use OpenRouter :free heavily, a one-time $10 balance unlock raises the daily cap from 50 RPD to 1,000 RPD.

🚀 High-TPM Fallback for Multi-File Agent Refactors

When free tiers throttle mid-diff, point Cline/Roo Code's second profile to a pay-as-you-go or self-hosted endpoint:

Disclosure: Novita AI and RunPod links above are referral/affiliate links that support FindAITier.