Alternatives to Sharbo

Fiddler AI is an enterprise AI control plane for agent observability, guardrails, and governance across the agentic lifecycle.

Modular humanoid robot platform for builders collecting data, training skills, and deploying autonomous robots.

Silico is Goodfire's interpretability agent for understanding, debugging, and controlling the internals of your AI models

Weights & Biases tracks ML experiments and traces LLM apps so teams can ship AI models faster

OpenCompass (司南) benchmarks LLMs, VLMs, and AI4S models against 100+ open evaluation datasets with published, dated leaderboards

MathArena is a free LLM math benchmark leaderboard scoring frontier models on ArXivLean proofs, BrokenArXiv, and ArXivMath competition

Open-source Unreal Engine plugin giving computer vision researchers synthetic images with pixel-perfect ground truth.

Cortex AI supplies real-workplace egocentric video and robot trajectory data for training embodied AI models.

AppWorld is a simulated-world benchmark for evaluating AI coding agents across 9 apps and 457 APIs.

Trace, evaluate, and iterate AI agents with Phoenix — open-source LLM observability you can self-host.

Agent and LLM observability from the LangChain team: trace, monitor, and evaluate agents in production, cloud, BYOC, or self-hosted.

Open-source LLM observability, prompt management, and evaluation for teams running AI agents in production.

Arena AI is a free LLM leaderboard where live head-to-head battles and community votes rank chat models, coding agents, and fullstack code.

Eval, tracing, red teaming, and governance for teams that need AI quality standardized across products.

ToolSpend tracks and forecasts AI API spend across OpenAI, Google AI, Azure, Bedrock, Anthropic, and Replicate in one dashboard.

Open-source workspace for building, evaluating, and deploying AI agents you talk to in Slack, Telegram, or a browser chat

Maxim AI simulates, evaluates, and observes AI agents in one workspace, with an open-source gateway (Bifrost) for routing and governance.

Open-source Python framework that replaces vibe checks with reproducible evaluation loops for RAG pipelines and AI agents.

Open-source Python framework for evaluating and monitoring LLMs, RAG apps, AI agents, and predictive ML models.

AI observability and eval engineering platform that turns offline evals into live production guardrails for agents and RAG systems.

Simulation-first evaluation and training infrastructure for AI agents, built on Digital World Models.

Developer observability platform that traces, replays, and debugs AI agent runs across OpenAI, CrewAI, Autogen and 400+ LLMs

Helicone is an AI gateway and LLM observability platform that routes, logs, and cost-tracks AI app traffic across 100+ models.

Open-source benchmark that scores AI agents on real, multi-step software-company work tasks.

Track token usage, spend, and quota limits across 41+ AI coding tools in one free desktop widget.

BALROG is an open ICLR-published benchmark that scores agentic LLM and VLM reasoning across seven procedurally generated games.

A Carnegie Mellon research benchmark of 910 visually grounded web tasks for multimodal browser agents, scored by execution rather than string matching.

ClawBench benchmarks AI browser agents on live websites with HTTP-interception scoring and LLM-judge grading.

Free daily LLM speed benchmarks — compare time-to-first-token and tokens-per-second across models and regions.

Local, MIT-licensed token and cost observability for AI coding agents — no SDK, no signup, nothing leaves your machine.