Skip to content

Llama

by Meta4 problems10 posts+3 this week

Problem posts per week

Problem posts per week18 posts, week of Aug 24 to week of Sep 21

Problems

ProblemThis week
Llama integrations break or are missingDev toolingLlama, Hugging Face, GitHub6d activeSeen on GitHub issuesUnverified+22 posts in the last 7 days, 0 the 7 days before (rising)
Deploying and serving models is harder than trainingMLOpsMLflow, GitHub, ChatGPT, Codex, Llama, Ollama7d activeSeen on GitHub issues, Hacker News, Reddit+11 post in the last 7 days, 2 the 7 days before (steady)
Running DeepSeek locally hits memory and speed limitsMLOpsDeepSeek, Hugging Face, Llama, llama.cpp, NVIDIA, GitHub2d activeSeen on GitHub issuesUnverified+11 post in the last 7 days, 1 the 7 days before (steady)
LLM responses are too slow for usersLLM appsCapacity, Claude, Claude Code, Llama, Ollama2d activeSeen on DEV, GitHub issuesUnverified+00 posts in the last 7 days, 1 the 7 days before (fading)
About

Problems here are about Llama or name it. It is tracked from the product list on the Method page. Website: meta.com.

Rising problems by email

Mondays: the problems in data, tech and AI that grew fastest that week.

Double opt-in. Unsubscribe any time.