synth-ai 0.5.4.dev20260120

Serverless Posttraining for Agents - Core AI functionality and tracing

Synth

Serverless Posttraining APIs for Developers

Average accuracy on LangProBe prompt optimization benchmarks.

Demo Notebooks (Colab)

• GEPA Banking77 Prompt Optimization
• GEPA Crafter VLM Verifier Optimization
• GraphGen Image Style Matching

Highlights

• 🎯 GEPA Prompt Optimization - Automatically improve prompts with evolutionary search. See 70%→95% accuracy gains on Banking77, +62% on critical game achievements
• 🔍 Zero-Shot Verifiers - Fast, accurate rubric-based evaluation with configurable scoring criteria
• 🧬 GraphGen - Train custom verifier graphs optimized for your specific workflows. Train custom pipelines for other tasks
• 🚀 No Code Changes - Wrap existing code in a FastAPI app and optimize via HTTP. Works with any language or framework
• ⚡️ Local Development - Run experiments locally with tunneled task apps. No cloud setup required
• 🗂️ Multi-Experiment Management - Track and compare prompts/models across runs with built-in experiment queues

Getting Started

uv add
uv run synth-ai tui

OpenCode Skills (Synth API)

The Synth-AI TUI integrates with OpenCode and ships a synth-api skill.

# List packaged skills shipped with synth-ai
uvx synth-ai skill list

uvx synth-ai skill install synth-api --dir ~/custom/opencode/skill

Testing

Run the TUI integration tests:

cd synth_ai/_tui
bun test

Synth is maintained by devs behind the MIPROv2 prompt optimizer.

Documentation

docs.usesynth.ai

Community

Join our Discord

GEPA Prompt Optimization (SDK)

Run GEPA prompt optimization programmatically:

import asyncio
import os
from synth_ai.sdk.api.train.prompt_learning import PromptLearningJob
from synth_ai.sdk.localapi import LocalAPIConfig, create_local_api

# Create a local task app: app = create_local_api(LocalAPIConfig(app_id="my_app", handler=my_handler))

# Create and submit a GEPA job
pl_job = PromptLearningJob.from_dict({
    "job_type": "prompt_learning",
    "config": {
        "prompt_learning": {
            "gepa": {
                "rollout": {"budget": 100},
                "population_size": 10,
                "generations": 5,
            }
        }
    },
    "task_app_id": "my_task_app",
})

pl_job.submit()
result = pl_job.stream_until_complete(timeout=3600.0)
print(f"Best score: {result.best_score}")

See the Banking77 demo notebook for a complete example with local task apps.

Zero-Shot Verifiers (SDK)

Run a built-in verifier graph with rubric criteria passed at runtime. See the Crafter VLM demo for verifier optimization:

import asyncio
import os
from synth_ai.sdk.graphs import VerifierClient

async def run_verifier():
    client = VerifierClient(
        base_url=os.environ["SYNTH_BACKEND_BASE"],
        api_key=os.environ["SYNTH_API_KEY"],
    )
    result = await client.evaluate(
        job_id="zero_shot_verifier_single",
        trace={"session_id": "s", "session_time_steps": []},
        rubric={
            "event": [{"id": "accuracy", "weight": 1.0, "description": "Correctness"}],
            "outcome": [{"id": "task_completion", "weight": 1.0, "description": "Completed task"}],
        },
        options={"event": True, "outcome": True, "model": "gpt-5-nano"},
        policy_name="my_policy",
        task_app_id="my_task",
    )
    return result

asyncio.run(run_verifier())

You can also call arbitrary graphs directly:

from synth_ai.sdk.graphs import GraphCompletionsClient

client = GraphCompletionsClient(base_url="https://api.usesynth.ai", api_key="...")
resp = await client.run(
    graph={"kind": "zero_shot", "verifier_shape": "mapreduce", "verifier_mode": "rubric"},
    input_data={"trace": {"session_id": "s", "session_time_steps": []}, "rubric": {"event": [], "outcome": []}},
)

GraphGen: Train Custom Verifier Graphs

Train custom verifier graphs using GraphGen. See the Image Style Matching demo for a complete GraphGen example:

from synth_ai.sdk.api.train.graphgen import GraphGenJob

# Train a verifier graph
verifier_job = GraphGenJob.from_dataset(
    dataset="verifier_dataset.json",
    graph_type="verifier",
    policy_models=["gpt-4.1"],
    proposer_effort="medium",  # Use "medium" (gpt-4.1) or "high" (gpt-5.2)
    rollout_budget=200,
)
verifier_job.submit()
result = verifier_job.stream_until_complete(timeout=3600.0)

# Run inference with trained verifier
verification = verifier_job.run_verifier(
    trace=my_trace,
    context={"rubric": my_rubric},
)
print(f"Score: {verification.score}, Reasoning: {verification.reasoning}")