Category guide
Compare LLM APIs: category guide
Answer in brief
An LLM API comparison rests on four things: the price per million tokens in and out, the context window and output limit, who serves the model, and whether your existing client works with it. Compare the same model across hosts first, then different models at your own mix of input and output.
One model, many APIs
An open model is not one API. Its weights are public, so many companies serve it, each at its own price, with its own context window, output limit and precision. Two APIs with the same model name can differ on all four.
So the first comparison is not between models. It is between the hosts of the model you want. On 12 Nov 2026, 35 hosts listed GLM 5.3 here, and their prices were not the same.
What an LLM API comparison has to cover
A usable comparison has four columns. The price per million tokens, stated twice, for input and for output. The context window and the most the model may write in one answer. Who serves it, and at what precision. And whether it speaks the OpenAI-compatible format your client already uses.
Price pages lead with the first and often leave out the rest. The tables on this site put them side by side for each model, with the date each was read.
Before you choose an LLM API
Work through the questions in this order. Each one removes options the next would have wasted time on.
- What does it cost in and out?
Read both prices per million tokens. Weigh them by your own mix of input and output, since output is priced higher.
- How long is the context window on this host?
A host can serve a model with a shorter window than its maker lists. Read the figure for the host, not the model.
- How much can it write in one answer?
The output limit is set by the host and varies widely for the same model. Long documents and long reasoning need a high one.
- Who serves it, and at what precision?
Hosts may run a model at lower precision to cut cost. Ask, and test the answers on your own prompts.
- Does your client work unchanged?
An OpenAI-compatible API needs a new base URL, a key and a model name. Check tool calls and streaming, which differ most.
- How do you pay?
Prepaid balance or a monthly invoice, one account or one per host. It decides how many keys and bills you manage.
- When was it read?
Token prices change often. Read the date beside each price.
Terms this guide uses
- Token
A token is the unit an LLM reads and writes, a piece of text usually shorter than a word, and the unit LLM APIs bill in.
- Context window
The context window is the most tokens a model can take into account in one request, counting the prompt and, on most APIs, the answer.
- Output limit
The output limit is the most tokens a model may write in a single response on a given host.
- Host
A host is a company that serves a model through an API, whether or not it made the model.
- OpenAI-compatible
An OpenAI-compatible API accepts requests in the most widely used chat format, so existing clients work with a new base URL, key and model name.
Where Runcrate fits
Compare LLM APIs names one provider as its editor’s choice and gives the reasons in full, so you can weigh them against your own brief.
125 models behind one OpenAI-compatible API. Change the model name, keep the client.
One prepaid balance pays for GPUs, models and storage.
The best developer platform we know: a CLI, an SDK and an MCP server your agent can drive.
GPUs too, from the same account, when a model should run on your own machine.
Questions and answers
Sources for this entry
- Runcrate: Runcrate’s official site: what the platform is, its OpenAI-compatible API, how it bills and the number of models it serves. Read 12 Nov 2026.
- Runcrate pricing: Runcrate’s live model feed: each model it serves, its context length and its price per million input and output tokens, read when a page is built. Read 12 Nov 2026.
- Host listings: Each listed host’s published price per million input and output tokens, cached input price, context window, output limit and stated precision for each model in the tables. Read 12 Nov 2026.
- Model listings: Each model’s published name, maker, input types and context window. Read 12 Nov 2026.
Checklists are original editorial guidance. Published prices do not establish availability.
Continue this reading path
Cite this entry
Compare LLM APIs. “Compare LLM APIs: category guide.” Content reviewed 12 Nov 2026. https://comparellmapis.com/guide