This app supports many inference services having an OpenAI compatible API. Below are some popular providers. All the information is for reference only. Please refer to the official docs of the providers for the most up-to-date information.
Models names listed here are just examples. You are welcome to experiment with those not listed. Although smaller models (<7B) tend to fail outputting in the desired format.
According to the docs from Google and community feedback, AI Studio offers a free tier giving okay
experience with Coreply. You may occasionally hit rate limits. Getting
...Unexpected JSON token at offset 0: Expected start of the object '{', but had '[' instead at path...
is likely caused by hitting rate limits. Please do not open issues for this.
API URL: https://generativelanguage.googleapis.com/v1beta/openai/
API Key: Grab it here
Model Name: gemini-2.0-flash, gemini-2.0-flash-lite, or gemini-2.5-flash-lite.
gemini-2.5-flash in AI Studio reasons by default. Therefore not recommended. gemma-3 family in
AI Studio doesn’t support system instructions. Thus not compatible.
According to the docs from Groq, Groq offers a free tier that likely works fine with Coreply. You may occasionally hit rate limits. Please do not open issues of getting errors if you are using the free tier, unless you are certain it’s not caused by rate limits.
API URL: https://api.groq.com/openai/v1/
API Key: Grab it here
Model Name: openai/gpt-oss-20b, llama-3.1-8b-instant, or llama-3.3-70b-versatile.
You are welcome to try other models listed in their docs.
According to the docs from Openrouter, some endpoints are free to use with strict limitations. It’s enough for testing but not recommended for regular use.
API URL: https://openrouter.ai/api/v1/
API Key: Grab it here
Model Name:
google/gemma-3-27b-it:free, meta-llama/llama-3.3-8b-instruct:free, and more.google/gemini-2.5-flash, google/gemini-2.5-flash-lite,
google/gemini-2.0-flash-001, google/gemini-2.0-flash-lite-001, openai/gpt-4.1-mini,
openai/gpt-4.1, mistralai/codestral-2508, openai/gpt-oss-20b, and more.You are welcome to try other models available on Openrouter.
It looks like OpenAI is very strict on free tier usage. It is not recommended to use Coreply with OpenAI without topping up any credits. Also note that having a subscription to ChatGPT ≠ topping up credits for API usage.
API URL: https://api.openai.com/v1/
API Key: Grab it here
Model Name: gpt-4.1-mini, gpt-4.1, gpt-4.1-nano, gpt-4o, or gpt-4o-mini.
gpt-5 family not recommended as they are reasoning models; mini variants are recommended for
the best cost-performance ratio.
Looks like Mistral offers a free tier, but needs more information about using it with Coreply. Please refer to their official docs.
Coreply supports codestral’s Fill-in-the-middle (FIM) API, which are low-latency and use fewer tokens. Besides that, chat completion API is also supported.
API URL: https://api.mistral.ai/v1/fim/
API Key: Grab it here
Model Name: codestral-latest
API URL: https://api.mistral.ai/v1/
API Key: Grab it here
Model Name: mistral-medium-2508 or mistral-small-2409
You are welcome to try other models from them.
Theoretically many OpenAI-compatible endpoints can be used. Find the API URL, API Key, and Model Name in the docs of the provider. The default system prompt works with mainstream models, but you can improve it if needed.