Alternatives to Matharena

OpenCompass (司南) benchmarks LLMs, VLMs, and AI4S models on dated, dimension-by-dimension leaderboards

Put a hard question to three frontier models — they answer blind, cross-examine each other, and hand back the one answer that survived, with a verdict.

Goodfire's Silico is an AI interpretability agent for reverse-engineering, debugging, and steering model internals.

Weights & Biases tracks ML experiments, versions models and datasets, and traces LLM apps — now inside CoreWeave Forge

Fiddler AI is an enterprise control plane for agent observability, inline guardrails, and auditable AI governance.

Free Hugging Face agent leaderboard ranking LLMs on planning, tool use, and multi-step agentic tasks

Open-source toolkit for benchmarking vision-language models across 80+ multimodal tasks, with rankings published on the Open VLM Leaderboard.

Sharbo builds a skill network for AI agents — encode expert workflows once and let every agent activate them autonomously.

OpenAI's December 2025 wet-lab evaluation where GPT-5 improved a real molecular cloning protocol 79-fold

Free AI benchmark where models build 3D voxel structures from text prompts and humans vote on spatial reasoning.

OpenJudge is an Apache 2.0 open-source AI evaluation framework shipping 50+ benchmark-validated graders for agents, LLMs, code, math, and multimodal models.

Helicone is an AI gateway and LLM observability platform that routes, caches, and logs traffic across 100+ model providers.

NannyML estimates your ML model's live performance without ground-truth labels and alerts you only when it actually matters

Open-source AI evaluation and observability for LLMs, RAG apps, AI agents, and ML models.

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

ViDoRe Benchmark: an open visual document retrieval benchmark and ColPali model suite for enterprise RAG evaluation.

OpenLIT delivers OpenTelemetry-native LLM tracing, evals, guardrails, and prompt management you self-host free under Apache 2.0.

Arena AI ranks chat models and coding agents on live head-to-head human votes and real agent sessions.

Token Monitor tracks AI coding token usage, cost, and quota limits across 44+ tools in one free desktop widget.

Compare 400+ AI models on one LLM Stats Score blending benchmarks, output speed and API pricing.

CodeClash benchmarks goal-oriented software engineering by making AI models build and evolve codebases to win 15-round arena tournaments.

QuickCompare replays your production LLM traffic on candidate models so you see which behaviours change before you switch.

Open-source PyTorch toolkit for feature attribution in sequence generation models

Open-source LLM observability: trace agents, manage prompts, and run evals on one MIT-licensed platform.

Open-source LLM observability and agent evaluation you can trace, self-host, and eval end to end.

Comet's Opik brings LLM observability and evaluation that turns agent traces into code fixes

AI quality workspace for LLM evaluation, production observability, red teaming, and governance

Open source AI engineering platform for agent and LLM tracing, evaluation, prompt management, and ML lifecycle work.

Maxim AI simulates, evaluates and observes AI agents — with Bifrost, its open-source LLM gateway, for routing and guardrails.

Braintrust traces agent runs, scores them with evals, and surfaces production failure patterns before they ship.