Keuxdi
@cody.bsky.social
Making things with computers. Run LLM’s directly on your phone 👉 esper-ai.com
I think I fixed sycophancy
August 14, 2025 at 4:05 PM
I think I fixed sycophancy
🧠 Qwen3 2507 thinking and instruct are now available in esper
💻 gpt-oss 20b is also available if you want to use esper on your M series mac
✨ Added Google Gemini to providers making it easier to setup your API key to access models, control reasoning with GPT-5, and much more
💻 gpt-oss 20b is also available if you want to use esper on your M series mac
✨ Added Google Gemini to providers making it easier to setup your API key to access models, control reasoning with GPT-5, and much more
August 13, 2025 at 7:00 PM
🧠 Qwen3 2507 thinking and instruct are now available in esper
💻 gpt-oss 20b is also available if you want to use esper on your M series mac
✨ Added Google Gemini to providers making it easier to setup your API key to access models, control reasoning with GPT-5, and much more
💻 gpt-oss 20b is also available if you want to use esper on your M series mac
✨ Added Google Gemini to providers making it easier to setup your API key to access models, control reasoning with GPT-5, and much more
SoftBank and OpenAI’s $500 Billion AI Project Struggles to Get Off Ground - WSJ
July 28, 2025 at 2:35 PM
SoftBank and OpenAI’s $500 Billion AI Project Struggles to Get Off Ground - WSJ
The entire Meta superintelligence team in 4 years after collecting their cheques
July 26, 2025 at 5:00 PM
The entire Meta superintelligence team in 4 years after collecting their cheques
This was clickbait. It's $1B now.
July 24, 2025 at 2:50 PM
This was clickbait. It's $1B now.
3 new models now available in esper!
@LiquidAI_'s LFM2 is designed specifically as an on-device model. It compares well to much larger models while being much faster and memory efficient.
I've also added Falcon H1 🇦🇪 and LG's hybrid reasoner EXAONE 🇰🇷
This brings the total to over 20 mo
@LiquidAI_'s LFM2 is designed specifically as an on-device model. It compares well to much larger models while being much faster and memory efficient.
I've also added Falcon H1 🇦🇪 and LG's hybrid reasoner EXAONE 🇰🇷
This brings the total to over 20 mo
July 23, 2025 at 4:00 PM
3 new models now available in esper!
@LiquidAI_'s LFM2 is designed specifically as an on-device model. It compares well to much larger models while being much faster and memory efficient.
I've also added Falcon H1 🇦🇪 and LG's hybrid reasoner EXAONE 🇰🇷
This brings the total to over 20 mo
@LiquidAI_'s LFM2 is designed specifically as an on-device model. It compares well to much larger models while being much faster and memory efficient.
I've also added Falcon H1 🇦🇪 and LG's hybrid reasoner EXAONE 🇰🇷
This brings the total to over 20 mo
"Meta’s superintelligence lab considers shift to closed AI model" - NYT
July 23, 2025 at 3:25 PM
"Meta’s superintelligence lab considers shift to closed AI model" - NYT
When it’s my turn during standup and I get to show off the button I made this week
July 22, 2025 at 2:45 PM
When it’s my turn during standup and I get to show off the button I made this week
Meta’s new super intelligence team wants to create a new closed model
July 21, 2025 at 2:35 PM
Meta’s new super intelligence team wants to create a new closed model
Openai ~4B open model
Grok 2/3 open sourced
Meta superintelligence lab edge model
Grok 2/3 open sourced
Meta superintelligence lab edge model
July 15, 2025 at 3:05 PM
Openai ~4B open model
Grok 2/3 open sourced
Meta superintelligence lab edge model
Grok 2/3 open sourced
Meta superintelligence lab edge model
The model weights are on huggingface
July 12, 2025 at 5:05 PM
The model weights are on huggingface
Local model summer
Everybody’s talking about it
Everybody’s talking about it
July 11, 2025 at 5:35 PM
Local model summer
Everybody’s talking about it
Everybody’s talking about it
*AGI being pursued by the least ethical companies and people*
Me:
Me:
July 10, 2025 at 2:35 PM
*AGI being pursued by the least ethical companies and people*
Me:
Me:
Prepping for Grok 4 and new OpenAI model release
July 9, 2025 at 5:35 PM
Prepping for Grok 4 and new OpenAI model release
On the eve of Grok 4, trained on the biggest cluster ever, we get a SOTA mobile model from Hugging Face.
Open weights, open data, open methods. Huge respect to the team for sharing everything.
Open weights, open data, open methods. Huge respect to the team for sharing everything.
July 9, 2025 at 3:39 AM
On the eve of Grok 4, trained on the biggest cluster ever, we get a SOTA mobile model from Hugging Face.
Open weights, open data, open methods. Huge respect to the team for sharing everything.
Open weights, open data, open methods. Huge respect to the team for sharing everything.
Where are the open Grok 2 weights you promised?
Elon:
Elon:
July 8, 2025 at 2:05 PM
Where are the open Grok 2 weights you promised?
Elon:
Elon:
Me: I literally have LLMs running on my phone
Everyone:
Everyone:
July 5, 2025 at 5:40 PM
Me: I literally have LLMs running on my phone
Everyone:
Everyone:
The first model under 10B with LMArena score of 1300+. Running on a phone.
July 4, 2025 at 2:20 PM
The first model under 10B with LMArena score of 1300+. Running on a phone.