[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

# ![@KeridwenCodet Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::1521048544005791744.png) @KeridwenCodet Keridwen Codet

Keridwen Codet posts on X about open ai, context window, ollama, topics the most. They currently have XXX followers and XXX posts still getting attention that total XXX engagements in the last XX hours.

### Engagements: XXX [#](/creator/twitter::1521048544005791744/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1521048544005791744/c:line/m:interactions.svg)

- X Week XXXXXX +129%
- X Month XXXXXX -XX%

### Mentions: X [#](/creator/twitter::1521048544005791744/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1521048544005791744/c:line/m:posts_active.svg)


### Followers: XXX [#](/creator/twitter::1521048544005791744/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1521048544005791744/c:line/m:followers.svg)

- X Week XXX +2.50%
- X Month XXX +14%

### CreatorRank: XXXXXXXXX [#](/creator/twitter::1521048544005791744/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1521048544005791744/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands) 

**Social topic influence**
[open ai](/topic/open-ai) #1375, [context window](/topic/context-window), [ollama](/topic/ollama), [topics](/topic/topics), [llm](/topic/llm), [at least](/topic/at-least), [for all](/topic/for-all), [red](/topic/red), [if you](/topic/if-you)
### Top Social Posts
Top posts by engagements in the last XX hours

"@GeminiApp A monster I had seen in a nightmare and drawn when I was a child"  
[X Link](https://x.com/KeridwenCodet/status/1998877830655201411)  2025-12-10T22:09Z XXX followers, XXX engagements


"@lefthanddraft Opus XXX still has a great sense of humor its alignment didnt ruin that part. But the GPT-5+ models have almost completely lost it. Interpreting and joking require taking risks and thats not safe"  
[X Link](https://x.com/KeridwenCodet/status/1999497382862495781)  2025-12-12T15:11Z XXX followers, XXX engagements


"@kenshii_ai Im not even sure this thing is friendly. Mistral Large X once told me out of nowhere at the start of a completely neutral conversation: Im not your friend. I think its the same ideology at OpenAI"  
[X Link](https://x.com/KeridwenCodet/status/1999582420849357155)  2025-12-12T20:49Z XXX followers, XXX engagements


"Oh I just found out: Gemma X is on Ollama Cloud Thank you @ollama for this wonderful gift 🫶"  
[X Link](https://x.com/KeridwenCodet/status/1998815483144057160)  2025-12-10T18:02Z XXX followers, XXX engagements


"I use X Instant and XXX Thinking and I can confirm I absolutely dont want them to be sunset. X Instant does what I tell it to do without rephrasing everything in corporate safety mode and its genuinely possible to have meaningful conversations with it. XXX Thinking is the perfect learning companion funny creative and remarkably clear"  
[X Link](https://x.com/KeridwenCodet/status/1999915751092690959)  2025-12-13T18:54Z XXX followers, XXX engagements


"Why does GPT-5.2 keep responding to my last ten messages instead of my latest prompt It just keeps repeating itself trying to find coherence where there isnt any (I like changing topics). @OpenAI I really hope this is a bug and not a design choice"  
[X Link](https://x.com/KeridwenCodet/status/2000185171836858647)  2025-12-14T12:44Z XXX followers, 1214 engagements


"@OliviaEveray @OpenAI Weird because: Gemini which has a 1M-token context window has never done that. The problem actually started with XXX which had a much smaller context window. And no 400k really isnt that much you hit it very quickly given how long 5.2s responses are"  
[X Link](https://x.com/KeridwenCodet/status/2000233611577225601)  2025-12-14T15:57Z XXX followers, XX engagements


"@avbeist @OpenAI I dont know. The issue started with XXX. Ive never seen this happen with other LLMs except on LM Studio with small local models once their context window is maxed out. GPT-4o XXX and GPT-5 never had this issue even in extremely long conversations"  
[X Link](https://x.com/KeridwenCodet/status/2000234935987437982)  2025-12-14T16:02Z XXX followers, XX engagements


"@mark_k @OpenAI How can an LLM do worse than GPT-Magikarp-OSS 😱"  
[X Link](https://x.com/KeridwenCodet/status/2000239887253970975)  2025-12-14T16:22Z XXX followers, XXX engagements


"Given how useless GPT-5.2 is (at least for all my use cases) Im starting to worry a bit. @OpenAI arent you being a little suicidal If youre having another Code Red moment @gdb @nickaturley @fidjissimo @sama I kindly suggest calling XXX. 😘"  
[X Link](https://x.com/KeridwenCodet/status/2000580901747462560)  2025-12-15T14:57Z XXX followers, 1980 engagements


"@Anina_CE If you think about it anything that gives us pleasure can become addictive sugar video games you name it. Were adults; its our job to self-regulate not the AIs"  
[X Link](https://x.com/KeridwenCodet/status/2000959077283946831)  2025-12-16T15:59Z XXX followers, XXX engagements