AI ModelBenchmarks
Find a model

Daily AI Model Scorecards

Browse today's winners, task leaders, and cost/latency tradeoffs across models with verified naming.

LiveSource log verified on 2026-08-19
Daily Model Eval Scorecard — 2026-04-11
Today's scorecard

Daily Model Eval Scorecard — 2026-04-11

AllCodingReasoningTool-useRAG / ResearchAgents

Eval Categories

💻

Coding

Bug fixes, refactors, API patches, diffs.

Current leader: Claude Fable 523 evals🧠

Reasoning

Decisions, tradeoffs, leadership calls.

Current leader: Claude Fable 520 evals🔧

Tool-use

Docs, CLI flows, integrations.

Current leader: GPT-5.520 evals📚

RAG / Research

Retrieval, synthesis, audit trails.

1 evals🤖

Agents

Multi-step plans, tool reliability.

2 evals⚡

Latency & Cost

Speed, spend, and throughput tradeoffs.

2 evals

Coding Scorecards

View all →

Bug fixes, diffs, refactors, and API patches tested across leading models.

Reasoning Scorecards

Decision quality under pressure, tradeoffs, and strategy calls.

Tool-use Scorecards

Docs-driven tasks, CLI accuracy, and security hygiene.

RAG / Research Scorecards

Retrieval accuracy, synthesis quality, and citation reliability.

Agent Reliability Scorecards

Multi-step task success, tool failures, and recovery behavior.

All Recent Scorecards

View all →