[GH-ISSUE #8972] Bad ouputs with Deepseek R1 api #15339

New Issue

2026-04-19T21:34:53-05:00

GiteaMirror commented

2026-04-19 21:34:53 -05:00

Originally created by @BenDes21 on GitHub (Jan 27, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/8972

Hi, sorry by advance Im a beginner.

I just install Open web UI on windows by pip install open-webui and launching the ui with open-webui serve then set my deep seek api to the settings > connections and simply add my api with the url https://api.deepseek.com / api_key

The model is working correctly on open web ui but I feel like the answers are less accurate / make more mistakes than the web app https://chat.deepseek.com/

Would like to know if I need to modify some settings or setup something else.

Thanks you

Originally created by @BenDes21 on GitHub (Jan 27, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/8972 Hi, sorry by advance Im a beginner. I just install Open web UI on windows by **pip install open-webui** and launching the ui with **open-webui serve** then set my deep seek api to the settings > connections and simply add my api with the url https://api.deepseek.com / api_key The model is working correctly on open web ui but I feel like the answers are less accurate / make more mistakes than the web app https://chat.deepseek.com/ Would like to know if I need to modify some settings or setup something else. Thanks you

GiteaMirror closed this issue

2026-04-19 21:34:54 -05:00

GiteaMirror referenced this issue

2026-04-19 23:19:03 -05:00

[GH-ISSUE #15339] issue: [v0.6.15] Open WebUI container uses VRAM at Idle #17537

GiteaMirror referenced this issue

2026-04-25 06:55:41 -05:00

[GH-ISSUE #15339] issue: [v0.6.15] Open WebUI container uses VRAM at Idle #33066

GiteaMirror referenced this issue

2026-05-05 18:57:08 -05:00

[GH-ISSUE #15339] issue: [v0.6.15] Open WebUI container uses VRAM at Idle #56203

GiteaMirror referenced this issue