proj: https://t.co/fCl3QBpnoE abs: https://t.co/i50YCzUKax repo: https://t.co/ml7dOTqYy1

@plainionist The industry has misled the public with terms like “In-Context Learning”, “Skills” or “Reasoning”. It’s trivial to prove no real learning is happening: the model’s weights are never changed. Remove the context, and the alleged “learning” disappears instantly. Puff.

Today we’re announcing Ternary Bonsai: Top intelligence at 1.58 bits Using ternary weights {-1, 0, +1}, we built a family of models that are 9x smaller than their 16-bit counterparts while outperforming most models in their respective parameter classes on standard benchmarks. We’re open-sourcing the models under the Apache 2.0 license in three sizes: 8B (1.75 GB), 4B (0.86 GB), and 1.7B (0.37 GB).

Squeeze Evolve: A Unified Framework for Verifier-Free Evolution Across AIME 2025, GPQA-Diamond, ARC-AGI-V2, MMMU-Pro, etc: - Up to ~3x API cost reduction - Up to ~10x increase in fixed-budget serving throughput https://t.co/MoTgGNAYns

Most serving stacks run FLUX.2 as four separate stages with Python overhead between each one. We collapsed all four into a single fused execution graph using MLIR-based compilation. On @AMD MI355X, that means a 3.8x speedup over torch.compile, 1024x1024 images in under 3.5 seconds, and a deployment container under 700MB. We ran the same pipeline on Blackwell, too. AMD delivers equivalent generation quality at a 5.5x lower cost. @clattner_llvm is presenting the full breakdown at AMD AI DevDay. Register: https://t.co/Pa1e36BTZn

We’ve been thinking a lot about scaling laws, wondering if there is a more effective way to scale FLOPs without increasing parameters. Turns out the answer is YES – by looping blocks of layers during training. We find that predictable scaling laws exist for layer looping, allowing us to use looping to achieve the quality of a Transformer twice the size. Our scaling laws suggest that for a fixed parameter budget, data and looping should be increased in tandem! 🧵👇

DMax Aggressive Parallel Decoding for dLLMs paper: https://t.co/y421NkegRD https://t.co/Y7Ut9Gxly8

Qwen3.6-35B-A3B just dropped. Red Hat AI has an NVFP4 quantized checkpoint ready. 35B params, 3B active, quantized with LLM Compressor. Preliminary GSM8K Platinum: 100.69% recovery (slightly above baseline). Early release. Let us know what you think! https://t.co/i5Fc4P7NVN

@DanielWulikk Have to think a bit about how to best visualize it, but if you are interested, I have a working from-scratch code implementation of Gemma 4 E2B in the meantime to see how per-layer embeddings are implemented: https://t.co/jyiq1vyJnH https://t.co/fVrSBWHNHl

We sat down with Kyle Caverly, an AI Performance Engineer on the MAX serve team, to walk through what actually happens inside an inference server from prompt to response. All the code discussed is open source. https://t.co/iwSQ4QA5F1

abs: https://t.co/tSLRNXAgY3 code: https://t.co/ldWVGmuC4O blog post: https://t.co/yI0NmdaKHm

📢 Super excited to announce Parcae! We've been thinking about scaling laws and the "right" way to get more FLOPs. Turns out layer looping - with the right parameterization - gives you a new axis to scale! Parcae matches Transformers 2x their size (w/ the same data), and outperforms prior formulations of looped models. But - you need the right parameterization to get these gains against strong Transformer baselines. Looped models are famously unstable to train, with tons of loss spikes and hyperparameter sensitivity. The main technical challenge with looped models is residual explosion - if you're passing the activations through the same layers over and over, some otherwise benign parameterizations cause huge instability. Our key idea: we can think of the residual stream of a model as a time-varying dynamical system - the same fundamentals behind SSMs like Mamba and S4. Then a few modest modifications to classic Transformers (stable diagonalization of injection params, LN before embeddings) can stabilize the looped models. The resulting models are more stable to train, but also reach higher quality. It's strong enough to start to derive new scaling laws. Classically - we know you need to scale parameters with data to be FLOP-optimal. With Parcae, we find a third axis - given fixed parameters, you additionally want to scale FLOPs by looping as you scale data. Super excited to see how these ideas hold, and what we can do with looped models! Check out @hayden_prairie's great explainer thread below, and see links for our paper, blog, and models. Joint w/ @zacknovack and @BergKirkpatrick, and a fun collab between @togethercompute and my lab at @ucsd_cse. Enjoy!

@trq212 This framing degrades technical literacy. Your prompt isn't "communication", it's tokenized, embedded as vectors, processed through frozen weights. No "agent" receives it. No bandwidth grows. Anthropic's own leaked system prompt: 16,739 words of context steering. That's engineering, not dialogue. Research shows anthropomorphic AI discourse creates self-fulfilling alignment degradation. You're not "talking to agents." You're structuring conditional probability queries. High-leverage? Yes. Interpersonal? Never. If the goal is public understanding, not marketing, stop treating inference as a “team member”. That's technically inaccurate. Which makes it dishonest the moment someone with enough expertise verifies it.

Also, how cool that this is so easy now? This was a few careful asks to Codex, which worked for ~128k tokens/1h to do everything - sourcing the data, embedding with clip (via @replicate), making an exploratory search tool for refining + filtering, and whipping up the final app.

Sharing my current setup to run Qwen3.6 locally in a good agentic setup (Pi + llama.cpp). Should give you a good overview of how good local agents are today: # Start llama.cpp server: llama-server \ -hf unsloth/Qwen3.6-35B-A3B-GGUF:Q4_K_XL \ --jinja \ --chat-template-kwargs '{"preserve_thinking":true}' \ --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0 # Configure Pi: { "providers": { "llama-cpp": { "baseUrl": "http://127.0.0.1:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "unsloth/Qwen3.6-35B-A3B-GGUF:Q4_K_XL" } ] } } }

dflash-mlx: DFlash speculative decoding, ported to Apple Silicon. Qwen3-4B at 186 tok/s on a MacBook. 4.6× faster than plain MLX-LM. Exact greedy decoding: output matches plain target decoding. https://t.co/VxfyworgAe

Introducing gyaradax 🐉: A JAX solver for local flux-tube gyrokinetics with custom CUDA kernels for acceleration. This entire code was vibecoded by @ggalletti_ and me in a month. Validated against GKW (CPU-only Fortran code) with 10x speedups. Details and code in the replies. https://t.co/22PrHjItR5

What if AI learned physics the way Newton did – by experiencing it? We built Sim2Reason: train LLMs inside virtual worlds governed by real physics laws, zero human annotation. Result: +5–10% improvement on International Physics Olympiad, zero-shot. 🧵 https://t.co/euXSyXmZVY

Earlier this month, Apple introduced Simple Self-Distillation: a fine-tuning method that improves models on coding tasks just by sampling from the model and training on its own outputs with plain cross-entropy and… it's already supported in TRL, built by @krasul. you can really feel the pace of development in the team 🐎 paper by @onloglogn, @richard_baihe, @UnderGroundJeg, Navdeep Jaitly, @trebolloc, @YizheZhangNLP at Apple 🍎 how it works: the model generates completions at a training-time temperature (T_train) with top_k/top_p truncation, then fine-tunes on them with plain cross-entropy. no labels or verifier needed you can try it right away with this ready-to-run example (Qwen3-4B on rStar-Coder): https://t.co/zizfISD6bq or benchmark a checkpoint with the eval script: https://t.co/mKlafTyKSe one neat insight from the paper: T_train and T_eval compose into an effective T_eff = T_train × T_eval, so a broad band of configs works well. even very noisy samples still help want to dig deeper? paper: https://t.co/aj1ZAcr8Mw trainer docs: https://t.co/TNVz93kZi9

+3 more

Cool to see that Meta conducted and published a pre-deployment investigation of Muse Spark behaviors like reward hacking, honesty, and evaluation awareness! https://t.co/i1Yy7HsEup

We comprehensively benchmarked Opus 4.7 on document understanding. We evaluated it through ParseBench - our comprehensive OCR benchmark for enterprise documents where we evaluate tables, text, charts, and visual grounding. The results 🧑‍🔬: - Opus 4.7 is a general improvement over Opus 4.6. It has gotten much better at charts compared to the previous iteration - Opus 4.7 is quite good at tables, though not quite as good as Gemini 3 flash - Opus 4.7 wins on content faithfulness across all techniques (including ours) - Using Opus 4.7 as an OCR solution is expensive at ~7c per page!! For comparison, our agentic mode is 1.25c and cost-effective is ~0.4c by default. Take a look at these results and more on ParseBench! https://t.co/tYiSOMbd6p

Codex just got a lot more powerful. Computer use, in-app browser, image generation and editing, 90+ new plugins to connect to everything, multi-terminal, SSH into devboxes, thread automations, rich document editing. Learns from experience and proactively suggestions work. And a ton more.

hermes @NousResearch agent with qwen3.5:35b-a3b on a 4090 is VERY good.. local models very impressive..

This 2-hour Stanford lecture breaks down how models like ChatGPT and Claude are actually built, clearer than what many people in top AI roles ever get exposed to. Save this and set aside two hours today. It might end up being the most valuable thing you learn all week. https://t.co/5u97uZCWxd

Ternary Bonsai: state-of-the-art intelligence at 1.58 bits. The models are so small they can even run locally in your browser on WebGPU! ⚡️ Here's the 8B version (just ~2GB in size) running at 60 tokens per second on my M4 Max. Try the demo out yourself! 👇

>We gave an AI agent an Apple Developer account, a Mac VM, and one task: build and publish an iOS app. It succeeded, at a cost of about $1,000. Great research on what it takes for an agent to do real world work. Some interesting areas of improvements:

Go forth and hack friends - Apache 2.0 license :)

MiniMax M2.7 is now on Together AI. Trained by letting it run its own RL loop, resulting in the highest open-source score on MLE Bench Lite. https://t.co/3E5mh9EDdz

To show off what you can do with @OpenAI Agent SDK + @modal, we built an ML research agent (inspired by @karpathy). It can: - Spin up GPU sandboxes of any shape - Run a pool of subagents - Persist memory - Snapshot state for fork/resume Here it is playing Parameter Golf: https://t.co/r7QhvNmdEq

We ran @AnthropicAI Claude Opus 4.7 (High) on APEX-SWE, our benchmark for real-world software engineering work. It scores 41.3% pass@1, placing 2nd on the leaderboard. It is only 0.2% away from GPT 5.3 Codex (High). https://t.co/4U6pKtrvI0

More details in Sayash's thread here: https://t.co/xcNemmsEs1

Gemma 4 31b scores 52.3% and is the strongest open model on on WeirdML, ahead of GLM 5 and gpt-oss-120b. This score is comparable to o3 and gemini 2.5 pro, and well ahead of qwen 3.5 27b at 39.5%. Gemma 4 is also significantly cheaper than other models with the same score. I ran this locally through ollama, with a 4-bit quant (q4_K_M). Full precision might score even better. The costs assume $0.14/$0.40/M.