All tags
Topic: "model-fine-tuning"
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.
OpenAI buys Jony Ive's io for $6.5b, LMArena lands $100m seed from a16z
gemini-2.5-pro gemini-diffusion openai lmarena a16z mistral-ai google google-deepmind multimodality reasoning code-generation math model-fine-tuning ai-assistants voice memory-optimization sundar_pichai
OpenAI confirmed a partnership with Jony Ive to develop consumer hardware. LMArena secured a $100 million seed round from a16z. Mistral launched a new code model fine-tune. Google DeepMind announced multiple updates at Google I/O 2024, including over a dozen new models and 20 AI products. Key highlights include the release of Gemini 2.5 Pro and Gemini Diffusion, featuring advanced multimodal reasoning, coding, and math capabilities, and integration of Gemini in Google Chrome as an AI browsing assistant. Deep Think enhanced reasoning mode and Project Astra improvements were also introduced, focusing on voice output, memory, and computer control for a universal AI assistant.
not much happened today
phi-4 phi-4-mini-reasoning qwen3-235b qwen3-moe-235b qwen3-moe-30b qwen3-dense-32b qwen3-dense-14b qwen3-dense-8b qwen3-dense-4b qwen3-dense-0.6b qwen2.5-omni-3b deepseek-prover-v2 llama llama-guard-4 prompt-guard-2 mimo-7b microsoft anthropic cursor alibaba togethercompute deepseek meta-ai-fair xiaomi openrouterai cohere reasoning model-fine-tuning model-evaluation benchmarking model-popularity open-source math model-scaling model-filtering jailbreak-prevention cline reach_vb vipulved akhaliq omarsar0 zhs05232838 huajian_xin mervenoyann karpathy random_walker sarahookr blancheminerva clefourrier
Microsoft released Phi-reasoning 4, a finetuned 14B reasoning model slightly behind QwQ but limited by data transparency and token efficiency issues. Anthropic introduced remote MCP server support and a 45-minute Research mode in Claude. Cursor published a model popularity list. Alibaba launched Qwen3-235B and other Qwen3 variants, highlighting budget-friendly coding and reasoning capabilities, with availability on Together AI API. Microsoft also released Phi-4-Mini-Reasoning with benchmark performance on AIME 2025 and OmniMath. DeepSeek announced DeepSeek-Prover V2 with state-of-the-art math problem solving, scaling to 671B parameters. Meta AI's Llama models hit 1.2 billion downloads, with new Llama Guard 4 and Prompt Guard 2 for input/output filtering and jailbreak prevention. Xiaomi released the open-source reasoning model MiMo-7B trained on 25 trillion tokens. Discussions on AI model evaluation highlighted issues with the LMArena leaderboard, data access biases favoring proprietary models, and challenges in maintaining fair benchmarking, with suggestions for alternatives like OpenRouterAI rankings. "LMArena slop and biased" and "61.3% of all data going to proprietary model providers" were noted concerns.
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.
Nvidia Minitron: LLM Pruning and Distillation updated for Llama 3.1
llama-3-1-8b llama-3-1 jamba-1.5 claude-3 dracarys-70b dracarys-72b mistral-nemo-minitron-8b mistral-7b nvidia meta-ai-fair ai21-labs anthropic hugging-face pruning knowledge-distillation weight-pruning activation-based-pruning width-pruning kl-divergence teacher-correction prompt-optimization multilinguality long-context mixture-of-experts model-fine-tuning
Nvidia and Meta researchers updated their Llama 3 results with a paper demonstrating the effectiveness of combining weight pruning and knowledge distillation to reduce training costs by training only the largest model from scratch and deriving smaller models via pruning and distillation. The process involves teacher correction, activation-based pruning (favoring width pruning), and retraining with distillation using KL Divergence loss, resulting in better-performing models at comparable sizes. However, distillation incurs some accuracy tradeoffs. Additionally, AI21 Labs launched Jamba 1.5, a hybrid SSM-Transformer MoE model with large context windows and multilingual support. Anthropic updated Claude 3 with LaTeX rendering and prompt caching. An open-source coding-focused LLM, Dracarys, was released in 70B and 72B sizes, showing improved coding performance. The Mistral Nemo Minitron 8B model outperforms Llama 3.1 8B and Mistral 7B on the Hugging Face leaderboard, highlighting pruning and distillation benefits. Research on prompt optimization reveals the complexity of prompt search spaces and the surprising effectiveness of simple algorithms like AutoPrompt/GCG.
AlphaProof + AlphaGeometry2 reach 1 point short of IMO Gold
gemini alphageometry-2 alphaproof llama-3-1-405b llama-3-70b llama-3-8b mistral-large-2 google-deepmind meta-ai-fair mistral-ai neurosymbolic-ai mathematical-reasoning synthetic-data knowledge-sharing model-fine-tuning alpha-zero multilinguality context-windows model-scaling benchmarking performance-comparison tim-gowers guillaume-lample osanseviero
Search+Verifier highlights advances in neurosymbolic AI during the 2024 Math Olympics. Google DeepMind's combination of AlphaProof and AlphaGeometry 2 solved four out of six IMO problems, with AlphaProof being a finetuned Gemini model using an AlphaZero approach, and AlphaGeometry 2 trained on significantly more synthetic data with a novel knowledge-sharing mechanism. Despite impressive results, human judges noted the AI required much longer time than human competitors. Meanwhile, Meta AI released Llama 3.1 with a 405B parameter model and smaller variants, and Mistral AI launched Mistral Large 2 with 123B parameters and 128k context windows, outperforming Llama 3.1 on coding tasks and multilingual benchmarks. This marks significant progress in AI mathematical reasoning, model scaling, and multilingual capabilities.
1/16/2024: TIES-Merging
mixtral-8x7b nous-hermes-2 frankendpo-4x7b-bf16 thebloke hugging-face nous-research togethercompute oak-ridge-national-laboratory vast-ai runpod mixture-of-experts random-gate-routing quantization gptq exl2-quants reinforcement-learning-from-human-feedback supercomputing trillion-parameter-models ghost-attention model-fine-tuning reward-models sanjiwatsuki superking__ mrdragonfox _dampf kaltcit rombodawg technotech
TheBloke's Discord community actively discusses Mixture of Experts (MoE) models, focusing on random gate routing layers for training and the challenges of immediate model use. There is a robust debate on quantization methods, comparing GPTQ and EXL2 quants, with EXL2 noted for faster execution on specialized hardware. A new model, Nous Hermes 2, based on Mixtral 8x7B and trained with RLHF, claims benchmark superiority but shows some inconsistencies. The Frontier supercomputer at Oak Ridge National Laboratory is highlighted for training a trillion-parameter LLM with 14TB RAM, sparking discussions on open-sourcing government-funded AI research. Additionally, the application of ghost attention in the academicat model is explored, with mixed reactions from the community. "Random gate layer is good for training but not for immediate use," and "EXL2 might offer faster execution on specialized hardware," are key insights shared.