Data Machina #262
Mistral NeMo 12B SOTA. Standford TexGrad. Patch-Level Training. Stanford STORM. State of Open AI. State of Txt2SQL. 450 Real World ML Systems. Convolutional Kernel Networks. EV-5 Universal Embeddings
datamachina.substack.com·platform·20 items·last fetched
Mistral NeMo 12B SOTA. Standford TexGrad. Patch-Level Training. Stanford STORM. State of Open AI. State of Txt2SQL. 450 Real World ML Systems. Convolutional Kernel Networks. EV-5 Universal Embeddings
Generative AI + Time-series Forecasting? The AI Agent Engineer. An Agentic Architecture? What's Arena Learning? AlphaFold3 Visualised. GraphRAG + Neo4j. Internet of Agents. Memory3 for LLMs.
Vision-Language Models Booming. PaliGemma. Phi-3 Vision. Florence-2. LLaVA-NeXT. ML in Video games. PCA in Latent Space. MosaicML Agents Framework. MoEs at Scale. GraphRAG. Image SSL on a Shoestring.
Prompt Engineering 2.0. Automated Prompt Optimisation. AI Scaling Myths. AGI World Models. What's an AI Agent? GenAI at LinkedIn. Failed AI Projects. Img2Txt2Txt Models. RAGFlow. JEPA Deep Dive.
AI and While You Were Out IRL. DeepSeek Coder v2. Hermes2+Theta Llama-3 70B. Unique 3D. AutoIF. Infinity Instruct. Florence. Claude 3.5 onnet. Claudette. Agile RL. TexGrad. PlanRAG.
Compound AI Systems, Txt2SQL & Data Agents. On Apple Intelligence Models. Lessons Building AI Agents. NVIDIA Nemotron-4 340B. New Memory Tuning. agentUniverse. Let's Reproduce GPT-2. Mixture of Agents
State Space Models (SSMs) An Alt to Transformers? Mamba-2. Chimera SSM Time-series. Audio Mamba. Sonic SSM Gen Voice. mamba.np. OSS Qwen-2 SOTA MLs. OSS LeRobot SOTA Robotics. Buffer of Thoughts.
New Trends in AI-RAG and Graphs. GRAG. GNN-RAG. Property Graph. Unified RAG+LangGraph. GenAI Mindset. Transformer Agents 2.0. Falcon 2.0 11B LLMS/ VLMS. ToonCrafter. MusePose. ColdFusion. SymbCoT.
State of AI Coding Agents. SWE-Agent. Amazon Q. Devin. OpenDevin. Devika. Blackbox AI. GPT-Engineer. ChatDev. KHOJ Personal AI Agents. Perplexica. CogVLM2. World Models.
The Google AI Blast. Gemini Pro 1.5. Gemini 1.5 Flash. PaliGemma. Project Astra. Delegation to AI Agents. NVIDIA ChatQA 1.5. Parler-TTS Mini:Expresso. DeepMind CAT3D. Meta AI Chameleon. KANs Explained
Diffusion, FM & Pre-Trained AI Models for Time-Series. TinyTimeMixers. MambaFormer. TimesFM. Frankenstein Prompts. BabyAGI. KANs Explained. GPT Researcher. xLSTM. Visualisation-of-Thought.
Six Nerdy AI Activities for the Long W/E. StoryDiffusion. AI Agents Stack. AI Town Game. Latest on In-Context Learning. KANs Alt to MLP. Amazon Q Assitant. Agentic RAG with llama3. WildChat Dataset.
Three New Powerful Open AI Models. Snowflake Artic. Apple OpenELM. Microsoft Phi-3. OpenVoicev2.Open-Sora. JAT Agent. GTE SOTA Embeddings. Maestro Subagents. Cohere RAG Toolkit. Diffusion GenAI Video.
Llama-3 Watershed Moment. Multi AI Agent Collaboration. AI Agents Planning. Idefics2-8B V-L Model. Google Gemini Cookbook. Quantisation Intro. torchtune. DeepMind Penzai. Youtube Commons Dataset.
GenAI Music. MusicGen. MusicFX. Stable Audio 2. Suno V3. Udio. Rerank3 Model. Parler TTS. nanoLLaVA VL Model. Text2SQL DuckDB-NSQL-7B. aiXcoder-7B
Jailbreaking AI Models is Easy. 4 New LLM Jailbreaking Methods. Mamba Model Primer. AI Agent Beats Humans on Kaggle. SWE-agent. RAGFlow. Stable Audio 2.0. VoiceCraft. AniPortrait. VAR SOTA ImageGen.
New Open Mixture-of-Experts Models. Jamba SSM-MoE. Qwen1.5-MoE-A2.7B. DBRX 132B MoE. frankenMoEs. AI Agentic Workflows. 1-bit ML Models. OpenDevin. AgentStudio.
Trends in Vision-Language Models. VideoAgent. MyVLM. ScreenAI. Evolutionary Model Merge. Embedding Quantisation. RAG 2.0 SOTA. LaVague Agent. Devika AI Engineer. Contextual Bandits. DenseFormer.
GenAI RAG Revisited. Command-R. RAFT. RAT. RAG + Knowledge Graph. Devin AI Engineer. KPU (Knowledge Processing Unit). Open-Sora GenAI Vid. AutoDev. DeepMind SIMA. DeepSeek-VL. Amazon Chronos Models.
AI Reasoning Like Humans. Self-Discover & Chain of Abstraction Reasoning. Claude 3 IQ Test. Neural Chess. FSDP + QLoRA. State of Competitive ML. Open Sora VideoGen.