Use Future AGI Models

What it is

Future AGI models are proprietary models built and optimized specifically for evaluation tasks — scoring, judging, and assessing AI outputs. They support text natively, with select models extending to images and audio. Each model offers a different trade-off between accuracy and latency, making them suited for different evaluation workloads across the platform and SDK.

---

Available models

• TURING_LARGE turing_large: Flagship evaluation model that delivers best-in-class accuracy across multimodal inputs (text, images, audio). Recommended when maximal precision outweighs latency constraints.

• TURING_SMALL turing_small: Compact variant that preserves high evaluation fidelity while lowering computational cost. Supports text and image evaluations.

• TURING_FLASH turing_flash: Latency-optimised version of TURING, providing high-accuracy assessments for text and image inputs with fast response times.

• PROTECT protect: Real-time guardrailing model for safety, policy compliance, and content-risk detection. Offers very low latency on text and audio streams and permits user-defined rule sets.

• PROTECT_FLASH protect_flash: Ultra-fast binary guardrail for text content. Designed for first-pass filtering where millisecond-level turnaround is critical.

---

Quick comparison

Model	Code	Inputs	Best for	Latency
TURING_LARGE	turing_large	Text, image, audio	Max accuracy, multimodal evals	Higher
TURING_SMALL	turing_small	Text, image	High fidelity, lower cost	Medium
TURING_FLASH	turing_flash	Text, image	Fast, high-accuracy evals	Low
PROTECT	protect	Text, audio	Safety, guardrails, user-defined rules	Low
PROTECT_FLASH	protect_flash	Text	First-pass binary filtering	Ultra-low

---

How to use

• In the UI — When you add or configure an evaluation (e.g. on a dataset or in a run test), choose Use Future AGI Models and pick one of the models above from the dropdown.

• In the SDK — Pass the model_name (e.g. turing_small, turing_flash, protect) in your evaluate() call. See Running your first eval for setup and usage.

  from fi.evals import Evaluator
  
  evaluator = Evaluator(fi_api_key="...", fi_secret_key="...")
  result = evaluator.evaluate(
      eval_templates="tone",
      inputs={"input": "Your text to evaluate."},
      model_name="turing_small",  # or turing_flash, turing_large, protect, protect_flash
  )

---

What you can do next