Use case ยท checked October 12, 2026

Cheapest LLM API for document summarization

Claude Haiku 5.5 and GPT-6 Luna tie as the cheapest at $16.00 a month. Our assumption: 20,000 documents a month, about 6,000 tokens in (roughly 4,500 words) and a 400-token summary out.

#ModelPer monthPer request
1Claude Haiku 5.5
Anthropic
$16.00$0.00080
2GPT-6 Luna
OpenAI
$16.00$0.00080
3DeepSeek V4.1 Flash
DeepSeek
$22.80$0.00114
4Gemini 3.1 Flash-Lite
Google
$42.00$0.00210
5Gemini 3.5 Flash-Lite
Google
$56.00$0.00280
6Mistral Large
Mistral
$72.00$0.00360
7DeepSeek V4 Pro
DeepSeek
$95.04$0.00475
8Gemini 3.8 Flash
Google
$120$0.00600
9Gemini 3.6 Flash
Google
$120$0.00600
10Grok Build 0.1
xAI
$136$0.00680
11Claude Haiku 4.5
Anthropic
$160$0.00800
12Grok 4.3
xAI
$170$0.00850
13Grok 4.7
xAI
$288$0.01440
14Claude Sonnet 5.5
Anthropic
$320$0.01600
15GPT-6.1 Sol
OpenAI
$320$0.01600
16GPT-5.3 Codex
OpenAI
$322$0.01610
17Gemini 3.1 Pro
Google
$336$0.01680
18Claude Opus 5.5
Anthropic
$640$0.03200
19Claude Fable 5.1
Anthropic
$1,600$0.08000
20GPT-6 Astra
OpenAI
$1,600$0.08000

Price only. Cheaper models are usually less capable, so test quality on your own data before switching.

Change the assumptions

Start from a use case

1,000 tokens is about 750 English words. Input includes your system prompt, chat history and any retrieved documents.

Claude Haiku 5.5
AnthropicCheapest
$16.00 /mo
GPT-6 Luna
OpenAICheapest
$16.00 /mo
$22.80 /mo
$42.00 /mo
$56.00 /mo
$72.00 /mo
$95.04 /mo
$120 /mo
$120 /mo
$136 /mo
$160 /mo
$170 /mo
$288 /mo
$320 /mo
$320 /mo
$322 /mo
$336 /mo
$640 /mo
$1,600 /mo
$1,600 /mo

Other use cases