News

Techmeme
techmeme.com > 260322 > p8

Tencent launches ClawBot, an OpenClaw-based agent integrated into WeChat, letting its 1B+ MAUs send and receive commands to interact with the AI agent via chat

4+ hour, 25+ min ago  (67+ words) Reuters: Tencent launches ClawBot, an OpenClaw-based agent integrated into WeChat, letting its 1B+ MAUs send and receive commands to interact with the AI agent via chat This is a Techmeme archive page. It shows how the site appeared at 4:50 PM ET,…...

Techmeme
techmeme.com > 260319 > p63

Microsoft releases MAI-Image-2, ranked #3 on the text-to-image Arena leaderboard behind models from Google and OpenAI, available in the MAI Playground

2+ day, 21+ hour ago  (69+ words) Jose Antonio Lanz / Decrypt: Microsoft releases MAI-Image-2, ranked #3 on the text-to-image Arena leaderboard behind models from Google and OpenAI, available in the MAI Playground This is a Techmeme archive page. It shows how the site appeared at 11:35 PM ET, March…...

Techmeme
techmeme.com > 260319 > p58

MiniMax releases M2.7, a proprietary “self-evolving” LLM that the company used to build, monitor, and optimize the model's own reinforcement learning harnesses

3+ day, 10+ min ago  (66+ words) Carl Franzen / VentureBeat: MiniMax releases M2.7, a proprietary "self-evolving" LLM that the company used to build, monitor, and optimize the model's own reinforcement learning harnesses This is a Techmeme archive page. It shows how the site appeared at 9:10 PM ET, March…...

Techmeme
techmeme.com > 260318 > p15

How AI's post-training process suppresses the creativity and whimsicality seen in earlier models, like GPT-2, leading to poor writing from many top AI models

4+ day, 14+ hour ago  (50+ words) Jasmine Sun / The Atlantic: How AI's post-training process suppresses the creativity and whimsicality seen in earlier models, like GPT-2, leading to poor writing from many top AI models This is a Techmeme archive page. It shows how the site appeared…...

Techmeme
techmeme.com > 260316 > p50

Mistral releases Small 4, its first model to unify reasoning, multimodal, and coding capabilities of its flagship Magistral, Pixtral, and Devstral models

5+ day, 21+ hour ago  (64+ words) Mistral AI: Mistral releases Small 4, its first model to unify reasoning, multimodal, and coding capabilities of its flagship Magistral, Pixtral, and Devstral models This is a Techmeme archive page. It shows how the site appeared at 12:30 AM ET, March 17, 2026. This…...

Techmeme
techmeme.com > 260311 > p6

Indian AI lab Lossfunk's prompting method lets LLMs generate Tulu language text without prior training; the method may expand to other low-resource languages

1+ week, 4+ day ago  (67+ words) The Economic Times: Indian AI lab Lossfunk's prompting method lets LLMs generate Tulu language text without prior training; the method may expand to other low-resource languages This is a Techmeme archive page. It shows how the site appeared at 2:00 AM…...

Techmeme
techmeme.com > 260308 > p5

A study finds LLMs from Anthropic, Google, OpenAI, and xAI can help with academic fraud, specifically helping non-researchers submit fabricated papers to arXiv

2+ week, 13+ hour ago  (47+ words) Elizabeth Gibney / Nature: A study finds LLMs from Anthropic, Google, OpenAI, and xAI can help with academic fraud, specifically helping non-researchers submit fabricated papers to arXiv This is a Techmeme archive page. It shows how the site appeared at 11:00 AM…...

Techmeme
techmeme.com > 260304 > p55

Microsoft releases Phi-4-reasoning-vision-15B, a 15B-parameter open-weight model it says matches larger systems while using far less compute and training data

2+ week, 3+ day ago  (49+ words) Michael Nu'ez / VentureBeat: Microsoft releases Phi-4-reasoning-vision-15B, a 15B-parameter open-weight model it says matches larger systems while using far less compute and training data This is a Techmeme archive page. It shows how the site appeared at 11:20 PM ET, March…...

Techmeme
techmeme.com > 260208 > p12

Q&A with mathematicians behind the “First Proof” experiment, which tests AI's mathematical competence on questions drawn from the authors' unpublished research

1+ mon, 2+ week ago  (46+ words) Siobhan Roberts / New York Times: Q&A with mathematicians behind the "First Proof" experiment, which tests AI's mathematical competence on questions drawn from the authors' unpublished research This is a Techmeme archive page. It shows how the site appeared at…...

Techmeme
techmeme.com > 260206 > p30

A look at the state of AI agents, the evolution of thinking models, the staggering need for inference compute in the coming years, automated research, and more

1+ mon, 2+ week ago  (51+ words) Eric Jang / Evjang.com: A look at the state of AI agents, the evolution of thinking models, the staggering need for inference compute in the coming years, automated research, and more This is a Techmeme archive page. It shows how…...