Agent platform inference vs Agent platform tuning
Agent platform inference is an open-source agent skill for ai models and llm tooling. Agent platform tuning is an open-source agent skill for ai models and llm tooling.
- Agent platform inference is best for run a test prompt against a gemini model with the genai sdk.
- Agent platform tuning is best for fine-tune a gemma or qwen model on my chat dataset.
- What it is
- Agent platform inference is an open-source agent skill for ai models and llm tooling.Agent platform tuning is an open-source agent skill for ai models and llm tooling.
- Best for
- Run a test prompt against a Gemini model with the GenAI SDKFine-tune a Gemma or Qwen model on my chat dataset
- Kind
- SkillSkill
- Pricing
- Open sourceOpen source
- Platforms
- Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, GooseClaude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, Goose
- GitHub stars
- 24.4K24.4K
- Last commit
- Nov 2026Nov 2026
- License
- Apache-2.0Apache-2.0
- Open source
- Good for
- Run a test prompt against a Gemini model with the GenAI SDK; Call a DeepSeek or Llama model through the OpenAI SDK; Query a tuned Gemini endpoint by its numeric IDFine-tune a Gemma or Qwen model on my chat dataset; Run supervised tuning on Gemini 2.5 Flash; Distill a larger model into a smaller student
What Agent platform inference does
Three model targets
Handles Gemini, open models as a service and custom endpoints.
Parameter grounding
Asks for model, SDK, project and region instead of defaulting.
Confirmation card
Lists parameters and waits for approval before spending quota.
Region probe
Checks Gemini availability with a live call to the exact region.
Endpoint call shapes
Chooses generateContent, chat completions or predict and the shared or dedicated host.
Error troubleshooting
Explains 429 DSQ, 400 validation, 404 and empty responses.
What Agent platform tuning does
Gated decision tree
Confirms project, region and model category before any action.
Dataset prep and validation
Converts to JSONL, validates schemas and confirms column mapping.
Validation split rule
Offers a 90/10 split and avoids the 80/20 split the service rejects.
Cost estimate
Calculates open-model tuning cost before asking for approval.
Open and Gemini jobs
Submits open-model jobs via script and Gemini jobs via the SDK, plus distillation.
Tuned model deployment
Registers the checkpoint and deploys it to a new endpoint after confirmation.
Agent platform inference vs Agent platform tuning: questions
Is Agent platform inference or Agent platform tuning cheaper?
Both are open source. Compare current plans on each official site.
What is the difference between Agent platform inference and Agent platform tuning?
Agent platform inference is an open-source agent skill for ai models and llm tooling. Agent platform tuning is an open-source agent skill for ai models and llm tooling.
