Best AI for summarising text

As of 2026-07-18, Anthropic: Claude Opus 4.8 (Fast) is the strongest model for this task in our daily benchmark, where over 350 AI models are evaluated and the best are presented.

A summary is only useful if it stays within the source and within the length you asked for. That is exactly where models diverge: many return a summary twice as long as requested, or add points that are not in the text.

For summaries we rank by how precisely the model follows instructions. Ranked by instruction score, from today’s benchmark.

Top choice

Anthropic: Claude Opus 4.8 (Fast)

anthropic

Score: 8.3/10

Language: 9.3/10

Speed: 131 t/s

$10.00

Runner-up

Anthropic: Claude Haiku 4.5

anthropic

Score: 8.0/10

Language: 9.3/10

Speed: 113 t/s

$1.00

Third choice

Mistral: Mistral Small 3

mistralai

Score: 7.8/10

Language: 8.7/10

Speed: 234 t/s

$0.05

Today’s ranking

Model	Provider	Score	Language	Speed	Input price/1M
Anthropic: Claude Opus 4.8 (Fast)	anthropic	8.3/10	9.3/10	131 t/s	$10.00
Anthropic: Claude Haiku 4.5	anthropic	8.0/10	9.3/10	113 t/s	$1.00
Mistral: Mistral Small 3	mistralai	7.8/10	8.7/10	234 t/s	$0.05
Perplexity: Sonar Pro	perplexity	7.7/10	9.0/10	68 t/s	$3.00
Cohere: Command R+ (08-2024)	cohere	7.5/10	9.3/10	160 t/s	$2.50
Meta: Llama 4 Maverick	meta-llama	7.2/10	8.0/10	176 t/s	$0.20
OpenAI: GPT-5.4 Nano	openai	7.1/10	8.0/10	157 t/s	$0.20
OpenAI: GPT-5.6 Sol Pro	openai	6.8/10	7.3/10	936 t/s	$5.00

Ask for a format, not just “a summary”

Say how long it should be, who it is for and what it will be used for: “five bullets for an executive making a decision” produces something entirely different from “summarise this”. It also helps to ask the model to separate what the text actually says from what it is inferring.

Check against the source

The most common failure in AI summaries is not outright fabrication but weighting: a side point gets promoted to the main conclusion. Always read the summary with the source alongside it when it will be used for anything that matters.

hvilkenAI is independent. We have no affiliate agreements, sponsored placements or commercial ties to the AI providers we test. Every recommendation rests on test results alone.

Frequently asked questions

Which AI is best at summarising?

The ranking above comes from today’s benchmark. We evaluate over 350 models and present the ones that follow instructions most precisely.

Can AI summarise an entire book?

Most models have enough context for long documents, but quality drops as the text gets very long. Split it into chapters and summarise stepwise for a better result.

Why is the summary longer than I asked for?

Models tend to elaborate. State a hard limit — “80 words maximum” — and repeat it at the end of the prompt.

Can I trust an AI summary?

As a starting point, yes. As the sole basis for a decision, no — verify the claims that actually matter against the original.