aiqbench · LLM benchmarks

Benchmarks

Per-axis LLM benchmarks. One leaderboard, one number, one source — no fabricated composites.

Sources: 8
Entries tracked: 22,318
Last sync: 34 min34 minutes ago

All sources syncing normally

Arena · Text· 54 minutes ago
Arena · WebDev· 54 minutes ago
Arena · Vision· 54 minutes ago
Arena · Agent· 54 minutes ago
GPQA Diamond· 34 minutes ago
SWE-bench· 34 minutes ago
Terminal-Bench· 34 minutes ago
ARC-AGI-2· 34 minutes ago

Arena · Text· 54 minutes ago
Arena · WebDev· 54 minutes ago
Arena · Vision· 54 minutes ago
Arena · Agent· 54 minutes ago
GPQA Diamond· 34 minutes ago
SWE-bench· 34 minutes ago
Terminal-Bench· 34 minutes ago
ARC-AGI-2· 34 minutes ago

Filter this table

Jump to a model's profile

Filter by org:

Released:AfterBefore

Fable 5.1

57.9%±3.4%

54.5%CI61.2%

Anthropicswing #167–#167

Fable 5

44.5%±3.8%

40.7%CI48.4%

Anthropicswing #280–#280

One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.

Axes1 on

Ranking on 1 axis — single-board order, no blending.

same = ranks models almost identically to an axis you already selected · close = nearly so · ? = too few shared models to tell.

Overall

Ability

Industry

Arena boards

Measured benchmarks

Agentic

Safety & factuality

Language

	Model	Org				± CI	Swing	Add to compare
4	Fable 5.1	Anthropic	—	—	57.9%	54.5%–61.2%	—
8	Fable 5	Anthropic	—	—	44.5%	40.7%–48.4%	—

Showing Terminal-Bench 4.0 exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.

Fable 5.1

Fable 5

The few things in AI and quantum that actually mattered this week — honestly summarised.