All tags
Model: "gpt-4-0613"
not much happened today
gpt-4-0613 gpt-3.5-turbo-0613 gpt-4o-2024-08-06 mistral-large-2 gpt4-turbo claude-3-opus idefics3-llama bigllama-3.1-1t-instruct llama-3-120b-instruct openai mistral-ai meta-ai-fair structured-outputs function-calling json-schema benchmarking multimodality context-windows model-scaling ai-hardware vision speech-processing robotics ai-regulation sama rohanpaul_ai corbtt guillaumelample mervenoyann maximelabonne aidan_mclau adcock_brett ylecun
OpenAI introduced structured outputs in their API with a new "strict" mode and a "response_format" parameter, supporting models like gpt-4-0613, gpt-3.5-turbo-0613, and the new gpt-4o-2024-08-06. They also halved the price of gpt-4o to $2.50 per million tokens. Mistral Large 2 outperforms gpt4-turbo and claude-3-opus on hard benchmarks and coding tasks. Idefics3-Llama offers multimodal capabilities with a 10k token context window. BigLlama-3.1-1T-Instruct is an upscaled version of llama-3-120b-instruct. New benchmark "big_model_smell" measures creativity and reliability. Figure 02 robot features advanced AI hardware with onboard vision language model, enhanced battery, and speech-to-speech reasoning. Yann LeCun expressed concerns about California's SB1047 regulation.
Music's Dall-E moment
griffin command-r-plus gpt-4-0613 gpt-4-0314 mistral-8x22b codegemma stable-diffusion-1.5 command-r gemini-1.5 google mistral-ai lmsys cohere model-architecture benchmarking open-source model-quantization memory-optimization inference-speed multimodality finetuning performance-optimization audio-processing andrej-karpathy
Google's Griffin architecture outperforms transformers with faster inference and lower memory usage on long contexts. Command R+ climbs to 6th place on the LMSYS Chatbot Arena leaderboard, surpassing GPT-4-0613 and GPT-4-0314. Mistral AI releases an open-source 8x22B model with a 64K context window and around 130B total parameters. Google open-sources CodeGemma models with pre-quantized 4-bit versions for faster downloads. Ella weights enhance Stable Diffusion 1.5 with LLM for semantic alignment. Unsloth enables 4x larger context windows and 80% memory reduction for finetuning. Andrej Karpathy releases LLMs implemented in pure C for potential performance gains. Command R+ runs in realtime on M2 Max MacBook using iMat q1 quantization. Cohere's Command R model offers low API costs and strong leaderboard performance. Gemini 1.5 impresses with audio capabilities recognizing speech tone and speaker identification from audio clips.
1/11/2024: Mixing Experts vs Merging Models
gpt-4-turbo gpt-4-0613 mixtral deepseekmoe phixtral deepseek-ai hugging-face nous-research teenage-engineering discord mixture-of-experts model-merging fine-tuning rag security discord-tos model-performance prompt-engineering function-calling semantic-analysis data-frameworks ash_prabaker shacrw teknium 0xevil everyoneisgross ldj pramod8481 mgreg_42266 georgejrjrjr kenakafrosty
18 guilds, 277 channels, and 1342 messages were analyzed with an estimated reading time saved of 187 minutes. The community switched to GPT-4 turbo and discussed the rise of Mixture of Experts (MoE) models like Mixtral, DeepSeekMOE, and Phixtral. Model merging techniques, including naive linear interpolation and "frankenmerges" by SOLAR and Goliath, are driving new performance gains on open leaderboards. Discussions in the Nous Research AI Discord covered topics such as AI playgrounds supporting prompt and RAG parameters, security concerns about third-party cloud usage, debates on Discord bots and TOS, skepticism about Teenage Engineering's cloud LLM, and performance differences between GPT-4 0613 and GPT-4 turbo. The community also explored fine-tuning strategies involving DPO, LoRA, and safetensors, integration of RAG with API calls, semantic differences between MoE and dense LLMs, and data frameworks like llama index and SciPhi-AI's synthesizer. Issues with anomalous characters in fine-tuning were also raised.