cd/sources/kagifeedback-auto-discovered· home› sources› Kagifeedback (auto-discovered)
cat /sources/kagifeedback-auto-discovered.feed | wc -l → 32

Kagifeedback (auto-discovered)

articles 32 domain kagifeedback.org → page 2/2 feed RSS
22:12
2026-09-10
kagifeedback.org
large-language-models

Add DeepSeek V4.1 Flash

A user identified as Mateo-Grgic posted a request to add DeepSeek V4.1 Flash, describing it as a new model that is better and cheaper than the existing DeepSeek V4.1 Flash. The post contains no furthe…

13:31
2026-09-10
kagifeedback.org
ai-products

MCP Support in Kagi Assistant

A feature request proposes adding Model Context Protocol (MCP) server support to Kagi Assistant, which would let the assistant interact with outside services through a plugin-style system where users …

00:28
2026-09-08
kagifeedback.org
ai-products

Implausible generation speed number in Assistant

A Kagi Assistant user reports that the displayed generation speed of 22 tokens per second contradicts the actual rate of 248 tokens per second calculated from 48,938 tokens generated in 197 seconds, s…

01:11
2026-09-05
kagifeedback.org
artificial-intelligence

Add GPT-6 Astra

OpenAI has released GPT-6 Astra, a new AI model, according to an announcement on the company's website.…

09:17
2026-09-03
kagifeedback.org
ai-products

Kagi Assistant Very Slow Today

Kagi Inc.'s AI assistant, Kagi Assistant, experienced severe slowdowns on March 5, 2025, with users reporting wait times of 1-2 minutes for basic queries that previously took 5 seconds, affecting both…

00:36
2026-09-03
kagifeedback.org
large-language-models

Add Claude Fable 5.1, Gemini 3.8 Flash

A user requested adding Claude Fable 5.1 and Gemini 3.8 Flash to an assistant, citing frequent use of Gemini 3.7 Flash and occasional use of Fable, and expressing a desire for stronger models.…

07:46
2026-08-31
kagifeedback.org
large-language-models

GLM 5.3-Flash, Qwen 3.8-Next

Kagi's documentation lists GLM 5.3-Flash as supported, but the model does not appear in all plans, and the page is inaccurate. A user suggests adding Qwen 3.8-Next/Flash and notes that the LLM privacy…

02:08
2026-08-31
kagifeedback.org
artificial-intelligence

Add apertus 1.5 70b model

Apertus AI released Apertus 1.5, a 70-billion-parameter open-weight and open-training-data model trained on a hydro-powered Swiss supercomputer, positioning it as an ethical alternative to frontier mo…

13:13
2026-08-29
kagifeedback.org
ai-products

Navigation menu Minimap in Assistent like in Vertex AI

A user requested a sidebar navigation minimap for Assistant threads, similar to Vertex AI's implementation, to allow jumping between messages in long Q&A sessions. The feature would improve navigation…

10:52
2026-08-29
kagifeedback.org
ai-products

Longer custom assistant instructions for custom assistants

OpenAI is considering raising the custom instruction limit for its AI assistants from the current 20,000 characters to 50,000 or 100,000 characters, according to a user request. The proposed increase …

21:41
2026-08-27
kagifeedback.org
artificial-intelligence

Audio has someone screaming on it?

A user on the Kagifeedback forum suggests that an AI-generated audio clip containing screaming is normal in gaming and amateur-made videos, and that the AI did nothing wrong by producing it. The user …

09:35
2026-08-27
kagifeedback.org
artificial-intelligence

Bug Translation Audio has someone screaming on it

A user on Kagifeedback.org reported that the AI voice model for Swedish in a game produces audio with someone screaming, but another user, laiz, suggests this is normal in gaming contexts and the AI d…

← prev page 2 / 2