The summarization leaderboard · for knowledge workers

Best for summarization

Decision area
Faithful, structured, no hallucinated bullets.
Editor picks
5 editor picks
Eligible models
14 eligible models
See raw /best
EDITOR'S CHOICEResearched 34d ago

Gemini 3.8 Flash

Google DeepMind · 1M context
Excellent

Low-cost models with long context windows for summarizing long documents.

Gemini 3.8 Flash takes text, image, video, audio and PDF input with a 1,048,576-token window and returns up to 65,536 output tokens. It is generally available on the Gemini API.

The numbers
$/1M out
$3.75
$0.75 input
Context
1M
max window
Pros
  • +1M-token context window
  • +Text, image, video, audio and PDF input
  • +Batch and flex options on the Gemini API
Cons
  • −Text output only

Also worth picking

The runners-up

ranked by editorial pick orderEditorial tiersExcellentStrongSolid
Anthropic · 1M
$0.50 / 1M out
Anthropic positions Haiku 5.5 for summaries and high-volume work: $0.10 / $0.50 per 1M tokens for prompts up to 100K tokens, and $0.50 / $2.50 above that, with a 1M window.
OpenAI · 1.05M
$0.50 / 1M out
OpenAI's GPT-6 Luna, for focused high-volume tasks, at $0.10 / $0.50 per 1M tokens with a 1.05M window.
Alibaba · 1M
$0.47 / 1M out
Qwen3.8-Omni-Flash takes text, image, audio and video input, with up to 991K tokens of input, at $0.15 / $0.47 per 1M tokens on Alibaba Cloud.
MiniMax · 1M
$1.20 / 1M out
MiniMax M3 uses MiniMax Sparse Attention for a 1M-token context, at $0.30 / $1.20 per 1M tokens on MiniMax's API.

Eligibility

14 models are eligible for this board

Eligibility means tagged with useCases: [summarization]. Pins must come from this pool.
All picks