i think in a world where mid and post training take equivalent or more compute (!!) than pretraining, this is less big a deal than what it used to mean in 2023
LLMS
-
Google Gemini 3.1 Flash TTS Brings Controllable Expressive Speech
By
–
🎙️AI voice just got a major upgrade.
— Futurepedia – Learn to Leverage AI (@futurepedia_io) 16 avril 2026
Google’s new Gemini 3.1 Flash TTS brings more controllable, expressive, high‑quality speech with simple prompt “audio tags” to steer tone, pacing, and style.
Rolling out in preview via Gemini API, Google AI Studio, Vertex AI and Google… pic.twitter.com/zPD3mbQJ9uAI voice just got a major upgrade. Google’s new Gemini 3.1 Flash TTS brings more controllable, expressive, high‑quality speech with simple prompt “audio tags” to steer tone, pacing, and style. Rolling out in preview via Gemini API, Google AI Studio, Vertex AI and Google
-
Local Qwen 35B Outperforms Claude Opus 4.7 on Benchmark
By
–
Shocking result on my pelican benchmark this morning, I got a better pelican from a 21GB local Qwen3.6-35B-A3B running on my laptop than I did from the new Opus 4.7! Qwen on the left, Opus on the right
-
Generative AI Tech Stack: Six Layers Powering Autonomous Agents
By
–
The Generative AI ecosystem is evolving into a full tech stack — powering autonomous AI agents.
From infrastructure and LLMs to RAG pipelines, agent behaviors and orchestration layers, this framework shows the 6 layers driving next-gen AI systems. Credit: @goyalshalini #AI -
Claude Opus 4.7 Now Powers Computer Orchestration Platform
By
–
Claude Opus 4.7 is now the default orchestration model powering Computer. It's also available for Max subscribers on Perplexity web, iOS, and Android.
-
MRCR Phase-Out: Shifting from Distractor-Based to Applied Long-Context
By
–
We kept MRCR in the system card for scientific honesty, but we've actually been phasing it out slowly. Two reasons: (1) it's built around stacking distractors to trick the model, which isn't how people actually use long context, and (2) we care more about applied long-context
-
Phasing Out MRCR: Shifting Focus from Distraction Tricks to Applied Long Context
By
–
We kept MRCR in the system card for scientific honesty, but we've actually been phasing it out slowly. Two reasons: (1) it's built around stacking distractors to trick the model, which isn't how people actually use long context, and (2) we care more about applied
-
Claude Opus 4.7 Launches on Replicate with Enhanced Vision
By
–
Claude Opus 4.7 is on Replicate. Anthropic's most capable model. Step-change in agentic coding. 3x better vision. 1M context. Try it now:
-
LangSmith Evaluators Hub: Centralized Management for AI Workspace
By
–
The Evaluators tab in LangSmith is a simple way to manage centrally. Our new hub surfaces all evaluators in your workspace, regardless of project. Build evaluators once, apply everywhere.
-
LangSmith Evaluation: New Reusable Evaluator Templates
By
–
New in LangSmith Evaluation:
— LangChain (@LangChain) 16 avril 2026
✅ Evaluator template library
✅ Reusable evaluators
Everything you need to know → https://t.co/OcHAwAdwAu pic.twitter.com/tE6RktsXxiNew in LangSmith Evaluation: Evaluator template library Reusable evaluators Everything you need to know → https://
langchain.com/blog/reusable-
langsmith-evaluator-templates?utm_source=x&utm_medium=social
…