llm-observability repositories
Trending repositories tagged llm-observability, ranked by trending score. Updated daily.
-
1
maximhq/bifrost
Open-source, high-performance AI gateway that works as your LLM gateway, MCP gateway and agent gateway. One OpenAI-compatible API for 10,000+ models, with failover, load balancing, semantic caching, virtual keys, budgets and MCP Code Mode. Works with any SDK or agent and adds under 200µs of overhead at 5k RPS.
GitHub · Go · 9,944 stars · 1,646 forks · +98 stars in 24 hours · Trending score 3.89
-
2
FailproofAI/failproofai
Observability and enforcement for AI agent harnesses. Capture every run and runtime reliability with policy enforcement.
GitHub · TypeScript · 10,018 stars · 580 forks · +99 stars in 24 hours · Trending score 3.83
-
3
langfuse/langfuse
🪢 Open source AI engineering platform: AI Gateway, agent observability, evals, and prompt/skill management, built on OpenTelemetry. One open platform for every AI call and every agent. Self-host or use Langfuse Cloud.
GitHub · TypeScript · 37,707 stars · 4,287 forks · +76 stars in 24 hours · Trending score 3.79
-
4
Continuum-AI-Corp/OrcaReplay
OrcaReplay — Time travel for AI agents. Record, replay, fork, and debug any agent run with any model. Built by the OrcaRouter.ai team.
GitHub · TypeScript · 1,088 stars · 176 forks · +49 stars in 24 hours · Trending score 3.74
-
5
scasella/claude-flightdeck
A Claude Code mod that puts a live agent dashboard in your terminal: context and cost, advisor timeline, every permission check, subagent cards and swimlanes.
GitHub · TypeScript · 334 stars · 52 forks · +77 stars in 24 hours · Trending score 3.28
-
6
comet-ml/opik
Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.
GitHub · Java · 23,481 stars · 1,968 forks · +24 stars in 24 hours · Trending score 2.91
-
7
jackulau/coding.sh
Open-source, local-first observability and evals for AI agents: see what Claude Code, Codex, Cursor and others did and what it cost.
GitHub · TypeScript · 149 stars · 6 forks · +21 stars in 24 hours · Trending score 2.75
-
8
traceroot-ai/traceroot
traceroot: the open-source platform that continuously improves your ai agents.
GitHub · TypeScript · 1,446 stars · 361 forks · +18 stars in 24 hours · Trending score 2.65
-
9
databufflabs/databuff
DataBuff is an AI-native APM built on Opentelemetry,with multi-agent troubleshooting out of the box.
GitHub · Java · 729 stars · 144 forks · +28 stars in 24 hours · Trending score 2.46
-
10
evilmartians/agent-prism
React components for visualizing traces from AI agents
GitHub · TypeScript · 409 stars · 25 forks · +11 stars in 24 hours · Trending score 1.81
-
11
future-agi/future-agi
Open-source, end-to-end platform for evaluating, observing, and improving LLM and AI agent applications. Tracing · Evals · Simulations · Datasets · Gateway · Guardrails. Self-hostable. Apache 2.0.
GitHub · Python · 2,285 stars · 725 forks · +4 stars in 24 hours · Trending score 1.78
-
12
openlit/openlit
Same agent, Smaller bill. The tools and infrastructure to get more work from your AI. Fewer tokens. Less compute. Lower cost.
GitHub · TypeScript · 3,094 stars · 495 forks · +3 stars in 24 hours · Trending score 1.72
-
13
taichuy/1flowbase
Build for agents — from the AI Gateway to the entire application stack.
GitHub · Rust · 296 stars · 24 forks · +9 stars in 24 hours · Trending score 1.69
-
14
Abilityai/trinity
Self-hosted AI Agents Platform for AI Native companies. Claude Code, Codex, Gemini, OpenCode agents. Apache 2.0.
GitHub · Python · 792 stars · 158 forks · +4 stars in 24 hours · Trending score 1.64
-
15
pydantic/logfire
AI observability platform for production LLM and agent systems.
GitHub · Python · 4,564 stars · 377 forks · +3 stars in 24 hours · Trending score 1.58
-
16
lmnr-ai/lmnr
Laminar - open-source observability platform purpose-built for AI agents. YC S24.
GitHub · TypeScript · 3,516 stars · 264 forks · +5 stars in 24 hours · Trending score 1.58
-
17
splunk/token-meter
Open-source, local-first AI coding agent usage and cost dashboard for Claude Code, Codex, Cursor, OpenCode, Kiro, and Pi.
GitHub · Python · 155 stars · 36 forks · +4 stars in 24 hours · Trending score 1.42
-
18
alibaba/loongsuite-pilot
Local-first telemetry collector for AI coding agents — unified OpenTelemetry events for Claude Code, Codex, Cursor and more. Token usage, cost, traces and security audit, exported anywhere.
GitHub · TypeScript · 265 stars · 67 forks · +2 stars in 24 hours · Trending score 1.36
-
19
dunetrace/dunetrace
The runtime reliability layer for AI agents.
GitHub · Python · 73 stars · 26 forks · +2 stars in 24 hours · Trending score 1.21
-
20
agenttrailhq/guard
Observability and guardrails for your ai agents. AgentTrail Guard traces your agent's execution and automatically compiles repeat failures into permanent guardrails.
GitHub · TypeScript · 14 stars · 3 forks · +2 stars in 24 hours · Trending score 0.91
-
21
duqaXxX/seedeep
What Claude Code is doing between your prompt and its answer, drawn live from the logs it already writes: every model call, every tool, each subagent on its own context window and its own model, and what the session produced. Local and read-only — your session content never leaves the machine.
GitHub · TypeScript · 49 stars · 3 forks · +3 stars in 24 hours · Trending score 0.91
-
22
anhermon/awesome-agent-observability
Curated list of LLM and AI agent observability and evaluation tools: tracing, evals, LLM-as-judge, guardrails, OpenTelemetry GenAI, MCP.
GitHub · Python · 66 stars · 36 forks · +1 stars in 24 hours · Trending score 0.85
-
23
mnfst/llm-gateway
Open Source LLM Gateway 🦚
GitHub · TypeScript · 7,639 stars · 531 forks · 0 stars in 24 hours · Trending score 0.83
-
24
ThinkWatchProject/ThinkWatch-Lite
Local gateway for Claude Code, Codex and other AI clients on macOS, Windows and Linux: switch upstreams without touching clients, keep API keys from relays, cut off malicious tool calls, trace every request.
GitHub · TypeScript · 946 stars · 8 forks · −30 stars in 24 hours · Trending score 0.75