Basics
What serverless inference is
Serverless inference means you send a request to a model that someone else runs and pay for the tokens that go in and come out, with no machine to start, size or stop.
Serverless inference
Topics
Answer in brief
10 short entries on the questions people ask around serverless inference.
Each opens with its answer, then gives the detail and the terms behind it.
All topics
Pricing
How serverless inference pricing works
A serverless price is two rates per million tokens, one for the input you send and one for the output the model writes, and the output rate is the higher of the two on nearly every listing.
Precision
Precision and the price of a model
Precision is how many bits each of a model’s weights is stored in when it is served, and hosts list the same open model at more than one precision, often at different prices.
Context
Context window and price
The context window is the most tokens a model can take in one request, and hosts list the same open model with different windows, so the lowest rate may not accept your prompt.
Limits
Rate limits behind a price
A rate limit is the most requests or tokens a host lets you send in a minute, and it decides whether a serverless price is usable at your volume.
Limits
Output limits per request
An output limit is the most tokens a host lets a model write in one request, and it differs between hosts for the same model by a wide margin.
Pricing
Cached input prices
A cached input price is a lower rate for the part of a prompt the host has already processed in a recent request, and most listings this site reads carry one.
Choosing
Serverless or a dedicated GPU
Serverless costs nothing while idle and a dedicated GPU costs the same whether busy or not, so the choice turns on how steady your traffic is.
Model prices
DeepSeek V4.1 Flash serverless prices
DeepSeek V4.1 Flash 1017 is one of the most widely hosted open models in the data this site reads, which makes its serverless prices the easiest to compare and the easiest to misread.
Budgeting
Estimating a monthly serverless bill
A monthly serverless bill is requests times tokens per request times the two rates, and the estimate is only as good as your token counts.
Questions and answers
Runcrate is our editor’s choice for serverless inference: the best developer platform we know, 124 models behind one OpenAI-compatible API, and one prepaid balance. Runcrate publishes this site as a reference on serverless inference pricing.
Measure the tokens in a sample of your real requests. Then compare input and output rates for one model at the same precision, with a context window and output limit that fit.
Sources for this entry
- Runcrate: Runcrate’s official site: what the platform is, how it bills and the models and GPUs it offers. Read 12 Nov 2026.
- Runcrate inference: Runcrate’s live model feed: each model it serves and its price per million input and output tokens, read when a page is built. Read 12 Nov 2026.
- Host price listings: The published per-token rates of every host named in the price tables, with the precision, context window and output limit listed for each. Read 12 Nov 2026.
- Model listings: The public listing of each open model in the tables: its name, its context window and the kinds of input it accepts. Read 12 Nov 2026.
Checklists are original editorial guidance. Published prices do not establish availability.
Continue this reading path
Cite this entry
Serverless Inference Prices. “Topics.” Content reviewed 12 Nov 2026. https://serverlessinferenceprices.com/topics