Best LLM for each use case
Best LLM for coding
The best LLMs for coding, ranked by public coding benchmarks (SWE-bench and similar). Cited scores, updated regularly.
Best LLM for reasoning
The best LLMs for reasoning, ranked by GPQA Diamond (graduate-level science questions). Cited scores, updated regularly.
Best LLM for agents
The best LLMs for agentic and tool-use workloads, ranked by Terminal-Bench 2.0. Cited scores.
Best LLM for long context
The LLMs with the largest context windows, for long-document and large-codebase workloads.
Cheapest LLM APIs
The cheapest LLM APIs, ranked by combined input + output price per 1M tokens, from official pricing pages.
Best vision LLM
The best multimodal LLMs that accept images, ranked by MMMU-Pro. Cited scores.