keiro 0.12.13

Keiro client — call the EB1 multi-model ensemble API.

Keiro

EB1 multi-model ensemble inference. Run multiple frontier models in parallel and synthesize the best response.

Quick start

pip install keiro
keiro setup

from keiro import models

print(models("eb1-preview", "What is machine learning?"))

Or from the command line:

keiro "What is machine learning?"

How it works

EB1 sends your prompt to multiple frontier models (Claude, GPT, Gemini) in parallel, then a judge synthesizes the strongest elements into a single response. The result is more accurate and more complete than any individual model.

Models

Model	Description
eb1-preview (default)	Adaptive GNN-routed ensemble
eb1-delta-preview	Adaptive ensemble with orchestration
eb1	Standard 5-model ensemble
eb1-pro	Extended 6-model ensemble
eb1-frontier	Highest quality, max reasoning
eb1-codex	Optimized for code and SWE tasks
eb1-fast	Low latency, lighter models
eb1-fast-preview	Adaptive routing, low latency
eb1-frontier-preview	Adaptive routing, max quality
claude-opus-4-6	Direct passthrough (no ensemble)
gpt-5.2	Direct passthrough

from keiro import models

# Default adaptive ensemble
answer = models("eb1-preview", "Solve this step by step: what is 23 * 47?")

# Max quality
answer = models("eb1-frontier", "Prove that sqrt(2) is irrational.")

# Low latency
answer = models("eb1-fast", "Summarize this in one sentence.")

# Direct passthrough to a single model
answer = models("claude-opus-4-6", "Write a haiku")

Prompt-first API

from keiro import models

# Structured response with usage metadata
reply = models.response("eb1-preview", "Explain quantum computing.")
print(reply.text)
print(reply.usage)

# Reusable model binding with fixed parameters
creative = models.instance("eb1-preview", temperature=0.8)
print(creative("Write a limerick about debugging."))

# Streaming
for chunk in models.stream("eb1-preview", "Draft a launch email."):
    print(chunk, end="")

Full client

from keiro import Client

client = Client()

# Chat completions API
response = client.chat(
    messages=[{"role": "user", "content": "Explain quantum computing."}],
    model="eb1-preview",
)
print(response["choices"][0]["message"]["content"])

# Rate limit visibility
print(client.rate_limits)
# RateLimitInfo(limit_requests=1000, remaining_requests=999, ...)

client.close()

CLI

keiro "What is ML?"                 # one-shot response
keiro                               # interactive REPL
keiro gui                           # local browser chat UI
keiro -m eb1-fast "Quick answer"    # specific model
echo context | keiro "Summarize"    # pipe context as input
keiro setup                         # configure credentials
keiro models                        # list available models

In the interactive REPL, streamed code fences render as numbered code blocks. Use /copy [n] to copy a block from the last assistant reply.

keiro gui opens the local browser chat UI in Chrome when available. If startup takes longer than expected, the CLI prints a manual URL and log path.

Configuration

Interactive setup (recommended):

keiro setup

This validates your API key against the gateway and saves credential metadata to ~/.keiro/credentials. Secret bytes are stored in owner-only sidecar files under ~/.keiro/secrets/, and the metadata file stores file:// references.

Explicit arguments:

from keiro import Client

client = Client(api_key="your-key", base_url="https://your-keiro-gateway.example")

API-key and endpoint precedence is explicit arguments, then credentials file. Runtime credential and gateway URL environment variables are ignored; run keiro setup or keiro endpoint local to update saved credentials.

Requirements

• Python 3.11+
• No GPU required (inference runs on hosted infrastructure)