PulseAugur
EN
LIVE 14:39:00

MiniStack 1.4.0 enables local Amazon Bedrock emulation with real LLM completions

MiniStack has released version 1.4.0, introducing local emulation of Amazon Bedrock services. This update allows developers to run Bedrock end-to-end on their local machines, initially providing deterministic mock responses for various model families like Anthropic, Titan, and Llama. A key feature is the ability to connect to local LLM backends such as Ollama via an environment variable, enabling real completions while maintaining the Bedrock API shape. AI

IMPACT Enables developers to test Bedrock integrations locally, reducing cloud costs and improving development speed.

RANK_REASON This is a tool release that enables local development for an existing cloud service, rather than a core AI model release or research.

Read on dev.to — LLM tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

MiniStack 1.4.0 enables local Amazon Bedrock emulation with real LLM completions

COVERAGE [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Nahuel Nucera ·

    Run Amazon Bedrock locally, with real completions from Ollama

    <p>MiniStack 1.4.0 ships four new services that emulate Amazon Bedrock end to end, and one env var that turns the mock into a real LLM backend.</p> <h3> The default: deterministic, shape-perfect mocks </h3> <p>Out of the box, Converse and InvokeModel return deterministic mock res…