workload

Answer questions over your docs

Retrieved chunks plus a question, short answer.

Retrieval pads the prompt, so caching the corpus changes the bill.

Input per request
6k
Output per request
500
Requests per month
20,000
Tokens per month
130M
Cheapest today
$2.58
Most expensive
$1,700

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$0.000129$2.58Open
GPT-OSS 20BOpenAIopen weights$0.018$0.090$0.000153$3.06Openaff
Qwen3.7 FlashAlibaba$0.030$0.130$0.000245$4.90Open
Llama 3.1 8BMetaopen weights$0.050$0.080$0.00034$6.80Openaff
DeepSeek V4 FlashDeepSeek$0.086$0.171$0.000601$12.03Open
Gemini 2.5 Flash-LiteGoogle$0.100$0.400$0.0008$16.00Open
GLM 5.3 FlashZ.ai$0.150$0.500$0.0012$23.00Open
Gemini 3.8 FlashGoogle$0.750$3.75$0.0064$128Open
Claude Haiku 4.5Anthropic$1.00$5.00$0.0085$170Openaff
Grok 4.3xAI$1.25$2.50$0.0088$175Open
GPT-5.1OpenAI$1.25$10.00$0.0125$250Open
Grok 4.6xAI$2.00$6.00$0.015$300Open
Claude Sonnet 5Anthropic$2.00$10.00$0.017$340Openaff
GPT-5.2 CodexOpenAI$1.75$14.00$0.0175$350Open
Gemini 3.1 ProGoogle$2.00$12.00$0.018$360Open
GPT-5.4OpenAI$2.50$15.00$0.0225$450Open
Kimi K3Moonshot$3.00$15.00$0.0255$510Open
Claude Opus 5.5Anthropic$4.00$20.00$0.034$680Openaff
GPT-5.5OpenAI$5.00$30.00$0.045$900Open
Claude Fable 5.1Anthropic$10.00$50.00$0.085$1,700Openaff
GPT-6 AstraOpenAI$10.00$50.00$0.085$1,700Open

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.