Post by Orange221 in v0.0.1 - Known Bugs, feedback, suggestions

Venus University » Community

Viewing post in v0.0.1 - Known Bugs, feedback, suggestions

← Return to game community

Orange22115 days ago(+1)

DeepSeek API gives HTTP 400 with OpenAI-compatible endpoint

Hi! I really like the game so far. I tried using my paid DeepSeek API key instead of Gemini.

Settings I used:

Endpoint: https://api.deepseek.com/v1/chat/completions
Model: deepseek-flash

When I click Test connection, I get: “Custom endpoint rejected the request (HTTP 400).”

DeepSeek's API is OpenAI-compatible, so I was wondering whether there is something in the request that isn't compatible with DeepSeek yet.

It would be great to have DeepSeek working as a custom provider, since I already have a paid DeepSeek API key and really like the game.

Thanks!

BadPersonPerry15 days ago(+1)

Yeah, that happens sometimes, either your reasoning is too high, which takes too long.
but then if you are using mimo 2.6 pro and glm 5.1. Reasoning has to be set to high or they won't work.
Deepseek 4 flash works, AFTER several retries.

Venus Dev15 days ago

Hey Orange, glad to hear you're enjoying the game. I'm not quite sure what's happening based on that message alone. Can you check the app.log? Maybe it has a more specific error in there

Orange22115 days ago(+1)

2026-09-23 16:44:53.437 main LOG [llm] → Custom endpoint models: https://api.deepseek.com/v1/models

2026-09-23 16:44:58.757 main LOG [llm] → Custom endpoint deepseek-flash (thinking=medium, tier=standard, maxTokens=65536, schema=connection_test)

2026-09-23 16:44:58.757 main LOG [llm] → system: (truncated, 20 chars)

2026-09-23 16:44:58.757 main LOG [llm] → user: Ping.

2026-09-23 16:44:59.089 main LOG [llm] ✕ Custom endpoint deepseek-flash HTTP 400 after 332ms

2026-09-23 16:44:59.089 main ERROR [ipc] llm:testWriter failed: {

code: 'LLM_REQUEST_REJECTED',

message: 'Custom endpoint rejected the request (HTTP 400).',

detail: 'invalid_request_error This response_format type is unavailable now'

}

Venus Dev15 days ago

So, apparently deepseek API does not support json_schema output type. That means that the API is wholly incompatible with our game unfortunately. If you want to use deepseek models, I tested using them through OpenRouter just now and it seemed fine. Sorry about this

Orange22115 days ago(+1)

ChatGBT made me a local proxy that converts the response format into a compatible one with deepseek, so I can actually use DeepSeek without needing to get it through OpenRoutner. So no worries.

Venus Dev15 days ago

Great! Glad it's working

itch.io

Viewing post in v0.0.1 - Known Bugs, feedback, suggestions