Preregistered Jacobian-lens reliability research on Gemma 4: 25k prompts, frozen probes, public traces, a prospective transfer miss, and an answer-identity confound.
-
Updated
Jul 19, 2026 - Python
#
llm-reliability
Here are 27 public repositories matching this topic...
Catch AI code hallucinations—fabricated packages, invented APIs, and contradicted behavior—deterministically, without trusting another model.
python ai developer-tools ai-agents hallucination-detection hallucination-mitigation llm-reliability hallucination-prevention ai-code-verification
-
Updated
Aug 30, 2026 - Python
Turn Chaos Into Structure. A Type-Safe AI Agent that extracts valid JSON from unstructured data using PydanticAI, FastHTML, and Gemini 2.5.
data-extraction fasthtml ai-agents unstructured-data type-safe-builder gemini-ai pydanticai self-healing-ai llm-reliability json-extraction
-
Updated
Jan 10, 2026 - Python
Interactive Phoenix LiveView demonstrations of the Crucible Framework - showcasing ensemble voting, request hedging, statistical analysis, and more with mock LLMs
visualization education demo machine-learning tutorial real-time elixir phoenix otp ai examples interactive beam phoenix-framework liveview ensemble-methods statistical-testing llm llm-reliability nshkr-crucible
-
Updated
Jul 16, 2026 - Elixir
Code for the ICML 2026 Main Track Paper "Cognitive Fatigue in Autoregressive Transformers: Formalization and Measurement"
nlp transformers icml model-interpretability large-language-models llm cognitive-fatigue llm-reliability fatigue-index icml-2026 production-llm inference-time-monitoring
-
Updated
Jul 16, 2026 - Jupyter Notebook
Collection of LLM failure modes used on failmodes.com
ai llms llm-security llm-evaluation llm-reliability ai-reliability llm-failure-modes llm-observation
-
Updated
Jun 14, 2026 - JavaScript
A Python to Prolog pipeline that bounds Elasticsearch LLM errors by grounding ES diagnostics analysis in verified metrics.
-
Updated
Jun 27, 2026 - Python
Reference implementation of CAAF — three-pillar agent framework with monotonic convergence.
constraint-satisfaction industrial-ai agent-framework llm-agents agentic-ai llm-reliability deterministic-ai
-
Updated
Apr 28, 2026 - Python
Extracts a driver profile from a call transcript behind a validating schema gate, then screens a load board before ranking by effective rate per mile.
-
Updated
Jul 24, 2026 - Python
Companion project for the TechnologyDig Academy tutorial on building reliable generative programs with Mellea.
-
Updated
Jul 19, 2026 - Python
TypeScript eval harness for measuring whether Grok answers stay grounded in source evidence
-
Updated
Jun 26, 2026 - TypeScript
Three small LLMs, one CPU, no cloud. Rigorous benchmarking, structured output validation, and head-to-head quality scoring via Ollama + FastAPI.
performance-benchmarking structured-output pydantic fastapi llm local-llm ollama llm-benchmarking llm-reliability
-
Updated
Aug 15, 2026 - Python
When the agent breaks, which layer dropped the ball? Operator-facing failure-attribution taxonomy for AI agent estates: ten-way dictionary, MAST + AgentRx crosswalks, postmortem template. Part of the Spine catalog.
taxonomy incident-response ai-agents postmortem root-cause-analysis agentops agent-architecture llm-reliability failure-attribution saasquach-ai-labs
-
Updated
Jul 17, 2026
Map where your bolted-on AI feature breaks before customers do. A free Claude Code tool: fragility map, reliability score across six dimensions, ranked gaps, and a 30-day plan. Built by a threat-intel practitioner.
hallucination rag prompt-engineering ai-testing llmops ai-quality llm-evaluation claude-code llm-reliability ai-evals ai-reliability claude-code-plugin
-
Updated
Jun 15, 2026
Public artifact bundle for the preprint 'Lightweight Evaluation and Operational Scorecards for Tool-Using AI Agents'
-
Updated
Jun 13, 2026 - Python
Make small local LLMs reliable at multi-step agent workflows. Constrained decoding + verify-before-commit: 96.5% end-to-end vs 2.0% unguarded on a 50-step run. Zero dependencies, runs on-prem.
python json-schema ai-agents on-premise structured-output agent-framework constrained-decoding llm llama-cpp vllm local-llm ollama llm-framework llm-agents small-language-models llm-reliability
-
Updated
Aug 25, 2026 - Python
CrucibleFramework: A scientific platform for LLM reliability research on the BEAM
documentation machine-learning elixir otp research ai reproducible-research beam reliability ai-research ensemble-methods statistical-testing research-framework experiment-framework llm llm-testing llm-reliability nshkr-crucible
-
Updated
Jul 16, 2026 - Elixir
Official implementation of Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation.
ai-safety nli rag cascade-detection langchain hallucination-detection agentic-ai llm-reliability multi-hop-qa
-
Updated
Aug 10, 2026 - Python
Profile README. AI engineer working on agentic LLM systems, deterministic guardrails, and how models fail in long interactions.
portfolio multi-agent-systems ai-engineer profile-readme llm-engineering agentic-ai langgraph-python llm-reliability
-
Updated
Sep 3, 2026
Reliability and hallucination mitigation research for tool-augmented legal AI agents using QC-Sentinel verification architecture.
-
Updated
Mar 6, 2026
Add this topic to your repo
To associate your repository with the llm-reliability topic, visit your repo's landing page and select "manage topics."