>65% of US physicians use OpenEvidence, with 27 million prompts in April https://
nbcnews.com/tech/tech-news
/openevidence-ai-doctor-medical-physician-login-app-what-npi-uptodate-rcna341064
…
LLMS
-
65% of US physicians use OpenEvidence with 27 million prompts in April
By
–
-
General AI models outperform specialized medical sources in study
By
–
For medical information, general AI frontier models (Google, OpenAI, Anthropic) outperformed specialized @EvidenceOpen and @UpToDate as assessed by 12 US clinicians, randomized and blinded to which model and extensive testing/benchmarks. This was not anticipated.
-
GPT 5.6 likely drops June 23, potentially harming Anthropic
By
–
I’m now 99.99% certain that GPT 5.6 will drop, in Codex and everywhere else, on June 23rd. This could end up being Anthropic’s biggest self own.
-
Kimi-K2.7: Major Coding & Reasoning Upgrade
By
–
Moonshot just released the Kimi-K2.7 code, a massive upgrade over Kimi-K2.6! Significant improvements over K2.6:
+21.8% on Kimi Code Bench v2
+11.0% on Program Bench
+31.5% on MLS Bench Lite
It also uses 30% fewer reasoning tokens, follows instructions better, and enhances long-horizon coding tasks. 6x -
Techniques for randomness and diversity in language model outputs
By
–
Summary of things: – turn up randomness and ban the most likely words (temperature + min-p + XTC sampling)
– ask for several different options at once, seed each with random constraints (verbalized sampling + entropy injection)
– give it a memory of what it's said and pick the -
Experimentation with Gemma 4 for creative fashion prompts
By
–
I experimented with using modifications of Gemma 4 to create creative prompts repeatedly. There are still a few quirks, but these are all results from the same simple query: "a dynamic fashion photo of a woman".
-
oMLX supports the standard HF cache directory
By
–
Great news: oMLX, by @jundotkim, now supports the standard HF cache model directory. Great MLX server for local AI! GG!
-
The author admits that ChatGPT and Claude are much smarter than him
By
–
ChatGPT and Claude are MUCH MUCH smarter than me in ALL areas. I am very amazed by the denial. People do not admit that AI is smarter than them. I accept the truth.
-
RL Scaling Laws for LLMs: From Pretraining to RL
By
–
Awesome blogpost! RL Scaling Laws for LLMs How scaling laws have evolved from pretraining to reinforcement learning… https://
cameronrwolfe.substack.com/p/rl-scaling-l
aws
… -
Claude Fable 5 claims #1 on DeepSWE Bench with 66% Pass@1
By
–
Claude Fable 5 has debuted on DeepSWE Bench with a 66% Pass@1, claiming the #1 spot and edging out GPT-5.5. The result reinforces a broader trend across recent coding benchmarks: strong raw performance combined with consistent reliability and efficiency in real-world software