workload

Summarize documents

Long input, short structured output.

The cheapest workload to move down-market, because input tokens carry it.

Input per request
8k
Output per request
600
Requests per month
2,000
Tokens per month
17.2M
Cheapest today
$0.340
Most expensive
$220

Most of the spread comes from output tokens on this workload. Applying a 60% prompt cache to the cheapest row moves it by $0 a month.

Open this in the calculator

All 21 models at list price, cheapest first.
ModelIn /1MOut /1MPer requestPer monthShareBuy
Mistral NemoMistralcheapestopen weights$0.019$0.030$0.00017$0.340Open
GPT-OSS 20BOpenAIopen weights$0.018$0.090$0.000198$0.396Openaff
Qwen3.7 FlashAlibaba$0.030$0.130$0.000318$0.636Open
Llama 3.1 8BMetaopen weights$0.050$0.080$0.000448$0.896Openaff
DeepSeek V4 FlashDeepSeek$0.086$0.171$0.000791$1.58Open
Gemini 2.5 Flash-LiteGoogle$0.100$0.400$0.001$2.08Open
GLM 5.3 FlashZ.ai$0.150$0.500$0.0015$3.00Open
Gemini 3.8 FlashGoogle$0.750$3.75$0.0083$16.50Open
Claude Haiku 4.5Anthropic$1.00$5.00$0.011$22.00Openaff
Grok 4.3xAI$1.25$2.50$0.0115$23.00Open
GPT-5.1OpenAI$1.25$10.00$0.016$32.00Open
Grok 4.6xAI$2.00$6.00$0.0196$39.20Open
Claude Sonnet 5Anthropic$2.00$10.00$0.022$44.00Openaff
GPT-5.2 CodexOpenAI$1.75$14.00$0.0224$44.80Open
Gemini 3.1 ProGoogle$2.00$12.00$0.0232$46.40Open
GPT-5.4OpenAI$2.50$15.00$0.029$58.00Open
Kimi K3Moonshot$3.00$15.00$0.033$66.00Open
Claude Opus 5.5Anthropic$4.00$20.00$0.044$88.00Openaff
GPT-5.5OpenAI$5.00$30.00$0.058$116Open
Claude Fable 5.1Anthropic$10.00$50.00$0.110$220Openaff
GPT-6 AstraOpenAI$10.00$50.00$0.110$220Open

Other workloads

Figures assume no retries, no failed calls, and no tool-use tokens. How the arithmetic works.

Prices read from 6 provider pricing pages on Sep 24, 2026. Some outbound links may be affiliate links. How that works.