PyPI / crates.io / RubyGems / Go index / NuGet
typellm 0.1.6(opens the publisher's site in a new tab)
Type-Safe Decoding for Autoregressive LLMs
Stories tagged ai.
PyPI / crates.io / RubyGems / Go index / NuGet
Type-Safe Decoding for Autoregressive LLMs
PyPI / crates.io / RubyGems / Go index / NuGet
UCCI: calibrated uncertainty for cost-optimal LLM cascade routing (token-margin signal, isotonic calibration, constrained threshold selection)
PyPI / crates.io / RubyGems / Go index / NuGet
The new inference engine for Computer Vision models
PyPI / crates.io / RubyGems / Go index / NuGet
Tolokaforge - LLM Tool-Use Benchmarking Harness

YourStory
The seven-year cloud infrastructure deal will support' Anthropic's growing CPU workloads and involves total capital expenditures of around $5.5 billion.
PyPI / crates.io / RubyGems / Go index / NuGet
MCP server for the Seren stack — the tool surface LLMs reach through.
PyPI / crates.io / RubyGems / Go index / NuGet
Three-tier LLM memory with consolidation for Seren. The right brain - episodic, time-tiered, dreaming the durable out of the fleeting. The Halls of Memory.
PyPI / crates.io / RubyGems / Go index / NuGet
Machine learning audio prediction experiments based on templates
PyPI / crates.io / RubyGems / Go index / NuGet
Zambo's 100+ tools for the OpenAI Agents SDK, with a verifiable receipt on every call.
PyPI / crates.io / RubyGems / Go index / NuGet
Ship hackathon products fast with secure-by-default auth, RBAC, PostgreSQL readiness, and built-in LLM tooling.
PyPI / crates.io / RubyGems / Go index / NuGet
Find out if your web page actually works. No tests written, no LLM.
PyPI / crates.io / RubyGems / Go index / NuGet
Lightweight multi-provider LLM SDK with streaming and tools
PyPI / crates.io / RubyGems / Go index / NuGet
Multi-backend Physics-Informed Neural Network (PINN) framework for solving, animating, and cross-validating ODEs/PDEs, with 50+ built-in equations, SDF-based geometry/CSG, and automatic architecture selection
PyPI / crates.io / RubyGems / Go index / NuGet
Test that a bot only ever does the job it was given — and never deletes, sends or approves the thing it must not. Record it doing the job once, then replay on every commit with zero LLM calls. For the agents and RPA already running real work in production: web, desktop, SAP, Citrix.

TNW
OpenAI is preparing to preview GPT-6 Cyber, a model built for cybersecurity work, in the coming weeks, Fortune reported on Thursday, citing multiple people familiar with the plans. Alongside the model, OpenAI plans to launch a new product that helps customers deploy it more securely and…
PyPI / crates.io / RubyGems / Go index / NuGet
Framework-neutral multi-provider LLM transport and Pydantic AI agent harness.
PyPI / crates.io / RubyGems / Go index / NuGet
Durable Python workspaces for stateful LLM agents
PyPI / crates.io / RubyGems / Go index / NuGet
Regression testing for LLM output, against a baseline committed in your own repo.
PyPI / crates.io / RubyGems / Go index / NuGet
Universal AI training & fine-tuning framework with native GPT-2 architecture, GPT-2 tokenizer support, smart SFT, and HF streaming.
PyPI / crates.io / RubyGems / Go index / NuGet
LLM observability SDK for Revefi - Traceloop-based monitoring with custom ingestor service
PyPI / crates.io / RubyGems / Go index / NuGet
Framework-neutral multi-provider LLM transport and Pydantic AI agent harness.
PyPI / crates.io / RubyGems / Go index / NuGet
AI-driven, self-healing web scraping: pluggable render backends + LLM-proposed extraction candidates validated and judged against real page content, never hand-rolled per-site parsers
PyPI / crates.io / RubyGems / Go index / NuGet
Tool framework for LLM agents -- routing, execution, and builtin tools
PyPI / crates.io / RubyGems / Go index / NuGet
Procedural memory for LLM agents -- skill definitions + invocation history