All tags
Company: "sakana-ai"
not much happened today
seedance-1.0 codex claude-code kling-2.1 veo-3 bytedance morph-labs huggingface deeplearning.ai figure-ai langchain sakana-ai video-generation autoformalization ai-assisted-coding api-design context-engineering reinforcement-learning ai-evals hypernetworks model-fine-tuning foundation-models andrew_ng hwchase17 adcock_brett clementdelangue akhaliq jxmnop hamelhusain sh_reya
Bytedance showcased an impressive state-of-the-art video generation model called Seedance 1.0 without releasing it, while Morph Labs announced Trinity, an autoformalization system for Lean. Huggingface Transformers deprecated Tensorflow/JAX support. Andrew Ng of DeepLearning.AI highlighted the rise of the GenAI Application Engineer role emphasizing skills in AI building blocks and AI-assisted coding tools like Codex and Claude Code. Engineering teams are increasingly testing API designs against LLMs for usability. Figure AI's CEO stressed speed as a key competitive advantage, and LangChain introduced the concept of Context Engineering for AI agents. Reinforcement learning on LLMs shows transformative potential, and the community values AI evals and data work. Sakana AI released Text-to-LoRA, a hypernetwork method for generating task-specific LoRA adapters from natural language, enabling efficient model customization. The video generation race heats up with Bytedance's Seed-based model praised for quality, challenging American labs, alongside models like Kling 2.1 and Veo 3.
lots of little things happened this week
llama-3-3-nemotron-super-49b-v1 claude anthropic nvidia sakana-ai meta-ai-fair reinforcement-learning reasoning benchmarks multi-turn-collaboration instruction-following dataset-release model-evaluation percy-liang
Anthropic introduced a novel 'think' tool enhancing instruction adherence and multi-step problem solving in agents, with combined reasoning and tool use demonstrated by Claude. NVIDIA's Llama-3.3-Nemotron-Super-49B-v1 ranked #14 on LMArena, noted for strong math reasoning and a 15M post-training dataset. Sakana AI launched a Sudoku-based reasoning benchmark to advance AI problem-solving capabilities. Meta AI released SWEET-RL, a reinforcement learning algorithm improving long-horizon multi-turn tasks by 6%, and introduced CollaborativeAgentBench, a benchmark for collaborative LLM agents working with humans on programming and design tasks. Percy Liang relaunched the HELM benchmark with 5 challenging datasets evaluating 22 top language models.
AI Engineer Summit Day 1
grok-3 o3-mini deepseek-r1 qwen-2.5-vl openai anthropic xai togethercompute alibaba sakana-ai benchmarking model-performance cuda model-training open-source debugging inference-speed batch-size reinforcement-learning aidan_mclau giffmana nrehiew_ teortaxestex epochairesearch andrew_n_carr borismpower yuhu_ai_
The AIE Summit in NYC highlighted key talks including Grace Isford's Trends Keynote, Neo4j/Pfizer's presentation, and OpenAI's first definition of Agents. Speakers announced $930 million in funding. On AI Twitter, discussions focused on Grok-3 and o3-mini models, with debates on performance and benchmarking, including Grok-3's record compute scale of 4e26 to 5e26 FLOP. The o3-mini model uncovered a critical CUDA kernel bug in Sakana AI's code. DeepSeek-R1 was promoted as an open-source alternative with notable training batch sizes. Additionally, Alibaba announced the Qwen 2.5-VL model release.
Reasoning Models are Near-Superhuman Coders (OpenAI IOI, Nvidia Kernels)
o3 o1 o3-mini deepseek-r1 qwen-2.5 openthinker openai nvidia ollama elevenlabs sakana-ai apple reinforcement-learning gpu-kernel-optimization fine-tuning knowledge-distillation scaling-laws chain-of-thought-reasoning model-accessibility alex-wei karpathy abacaj awnihannun
o3 model achieved a gold medal at the 2024 IOI and ranks in the 99.8 percentile on Codeforces, outperforming most humans with reinforcement learning (RL) methods proving superior to inductive bias approaches. Nvidia's DeepSeek-R1 autonomously generates GPU kernels that surpass some expert-engineered kernels, showcasing simple yet effective AI-driven optimization. OpenAI updated o1 and o3-mini models to support file and image uploads in ChatGPT and released DeepResearch, a powerful research assistant based on the o3 model with RL for deep chain-of-thought reasoning. Ollama introduced OpenThinker models fine-tuned from Qwen2.5, outperforming some DeepSeek-R1 distillation models. ElevenLabs grew into a $3.3 billion company specializing in AI voice synthesis without open-sourcing their technology. Research highlights include Sakana AI Labs' TAID knowledge distillation method receiving a Spotlight at ICLR 2025, and Apple's work on scaling laws for mixture-of-experts (MoEs). The importance of open-source AI for scientific discovery was also emphasized.
Mistral Small 3 24B and Tulu 3 405B
mistral-small-3 tulu-3-405b llama-3 tiny-swallow-1.5b qwen-2.5-max deepseek-v3 claude-3.5-sonnet gemini-1.5-pro gpt4o-mini llama-3-3-70b mistral-ai ai2 sakana-ai alibaba_qwen deepseek ollama llamaindex reinforcement-learning model-fine-tuning local-inference model-performance model-optimization on-device-ai instruction-following api training-data natural-language-processing clementdelangue dchaplot reach_vb
Mistral AI released Mistral Small 3, a 24B parameter model optimized for local inference with low latency and 81% accuracy on MMLU, competing with Llama 3.3 70B, Qwen-2.5 32B, and GPT4o-mini. AI2 released Tülu 3 405B, a large finetuned model of Llama 3 using Reinforcement Learning from Verifiable Rewards (RVLR), competitive with DeepSeek v3. Sakana AI launched TinySwallow-1.5B, a Japanese language model using TAID for on-device use. Alibaba_Qwen released Qwen 2.5 Max, trained on 20 trillion tokens, with performance comparable to DeepSeek V3, Claude 3.5 Sonnet, and Gemini 1.5 Pro, and updated API pricing. These releases highlight advances in open models, efficient inference, and reinforcement learning techniques.
$1150m for SSI, Sakana, You.com + Claude 500m context
olmo llama2-13b-chat claude claude-3.5-sonnet safe-superintelligence sakana-ai you-com perplexity-ai anthropic ai2 mixture-of-experts model-architecture model-training gpu-costs retrieval-augmented-generation video-generation ai-alignment enterprise-ai agentic-ai command-and-control ilya-sutskever mervenoyann yuchenj_uw rohanpaul_ai ctojunior omarsar0
Safe Superintelligence raised $1 billion at a $5 billion valuation, focusing on safety and search approaches as hinted by Ilya Sutskever. Sakana AI secured a $100 million Series A funding round, emphasizing nature-inspired collective intelligence. You.com pivoted to a ChatGPT-like productivity agent after a $50 million Series B round, while Perplexity AI raised over $250 million this summer. Anthropic launched Claude for Enterprise with a 500 million token context window. AI2 released a 64-expert Mixture-of-Experts (MoE) model called OLMo, outperforming Llama2-13B-Chat. Key AI research trends include efficient MoE architectures, challenges in AI alignment and GPU costs, and emerging AI agents for autonomous tasks. Innovations in AI development feature command and control for video generation, Retrieval-Augmented Generation (RAG) efficiency, and GitHub integration under Anthropic's Enterprise plan. "Our logo is meant to invoke the idea of a school of fish coming together and forming a coherent entity from simple rules as we want to make use of ideas from nature such as evolution and collective intelligence in our research."
The DSPy Roadmap
dspy litel-lm gemini chatgpt-4o grok-2 hermes-3 databricks mit google openai x-ai nous-research astribot apple sakana-ai model-optimization fine-tuning optimizers interactive-optimization robotics autonomous-systems voice image-generation open-source-models scientific-research streaming caching omar-khattab giffmana
Omar Khattab announced joining Databricks before his MIT professorship and outlined the roadmap for DSPy 2.5 and 3.0+, focusing on improving core components like LMs, signatures, optimizers, and assertions with features such as adopting LiteLLM to reduce code and enhance caching and streaming. The roadmap also includes developing more accurate, cost-effective optimizers, building tutorials, and enabling interactive optimization tracking. On AI Twitter, Google launched Gemini Live, a mobile conversational AI with voice and 10 voices, alongside Pixel Buds Pro 2 with a custom Tensor A1 chip. OpenAI updated ChatGPT-4o, reclaiming the top spot on LMSYS Arena. xAI released Grok-2 in beta, achieving SOTA in image generation with FLUX 1. Nous Research released open-source Hermes 3 models in 8B, 70B, and 405B sizes, with the 405B model achieving SOTA. Robotics updates include Astribot's humanoid robot and Apple's tabletop robot with Siri voice commands. Sakana AI introduced "The AI Scientist," an autonomous AI research system.
Hybrid SSM/Transformers > Pure SSMs/Pure Transformers
mamba-2-hybrid gpt-4 qwen-72b table-llava-7b nvidia lamini-ai sakana-ai luma-labs mixture-of-experts benchmarking fine-tuning multimodality text-to-video model-performance memory-optimization preference-optimization video-understanding multimodal-tables bryan-catanzaro bindureddy ylecun ctnzr corbtt realsharonzhou andrew-n-carr karpathy _akhaliq omarsar0
NVIDIA's Bryan Catanzaro highlights a new paper on Mamba models, showing that mixing Mamba and Transformer blocks outperforms either alone, with optimal attention below 20%. Mixture-of-Agents (MoA) architecture improves LLM generation quality, scoring 65.1% on AlpacaEval 2.0 versus GPT-4 Omni's 57.5%. The LiveBench AI benchmark evaluates reasoning, coding, writing, and data analysis. A hybrid Mamba-2-Hybrid model with 7% attention surpasses a Transformer on MMLU accuracy, jumping from 50% to 53.6%. GPT-4 performs better at temperature=1. Qwen 72B leads open-source models on LiveBench AI. LaminiAI Memory Tuning achieves 95% accuracy on a SQL agent task, improving over instruction fine-tuning. Sakana AI Lab uses evolutionary strategies for preference optimization. Luma Labs Dream Machine demonstrates advanced text-to-video generation. The MMWorld benchmark evaluates multimodal video understanding, and Table-LLaVa 7B competes with GPT-4V on multimodal table tasks.