Choose a model on evidence, not on a launch post.

Rule: 340 of 1100 catalog models — those with at least one of an overall score, a token price, or a throughput measurement.

Evidence key: Observed

Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.

All models

Model	Access	Score	$/1M in	$/1M out	TTFT	Throughput
Claude Opus 5.5Anthropic	Proprietary	86.68	$4.00	$20.00	573.15 s	95 tok/s
GPT-6.1 SolOpenAI	Proprietary	84.22	$2.00	$10.00	Unavailable	38.5 tok/s
GPT-6 AstraOpenAI	Proprietary	84.04	$10.00	$50.00	265.1 s	66 tok/s
Claude Fable 5.1Anthropic	Proprietary	83.22	$10.00	$50.00	206.2 s	69 tok/s
Gemini 4 ArgonGoogle	Proprietary	82.89	$2.00	$10.00	47.97 s	83 tok/s
Claude Sonnet 5.5Anthropic	Proprietary	82.24	$2.00	$10.00	394.99 s	136 tok/s