siegevjorn

• Qwen 3.6 coding choice–27B vs 35B quants

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 96 comments

• Qwen 35b a3b surprises me

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 38 comments

• Which inference engine to choose for mlx?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 17 comments

• "DeepSeek produced a model close to the performance of US models 7-10 months older, for a good deal less cost (but NOT anywhere near the ratios people have suggested)" says Anthropic's CEO

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 446 comments

• What agentic cli do you use for local models ?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 12 comments

• Llama.cpp: vlm access via llama-server causes cuda OOM error after processing 15k images.

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 2 comments

• llm-compressor: vLLM AWQ quant with multiple GPUs keep causing errors

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 2 comments

• Claude code Max vs. Mac Studio M4 Max 128gb running open code

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 40 comments

• Fine-tuning llms on dgx spark from nvidia webpage

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 2 comments

• Analyzing email thread: hallucination

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 10 comments

• M2 to PCIEx16 adaptor safety

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 2 comments

• 'My Productivity Is At Zero': Meme Frenzy On Social Media As ChatGPT Goes Down Globally

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 6 comments

• Curious how AMD (Radeon) GPUs can handle LLMs

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 34 comments

• Dual 3090 configurations—Are used 3090s reliable enough?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 3 comments

• Myth about nvlink

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 10 comments

• Why is it hard to find LLM size that fits consumer-grade GPUs?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 57 comments

• Everyone and their mother knows about DeepSeek

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 360 comments

• Open-R1: a fully open reproduction of DeepSeek-R1 from huggingface

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 8 comments

• Laptop inference speed on Llama 3.3 70B

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 75 comments

• Red Hat Announces Definitive Agreement to Acquire Neural Magic (vLLM)

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 50 comments

• GPU poor's dilemma: 3060 12GB vs. 4060 Ti 16GB

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 27 comments

• Boy is 5090 beautiful

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 1 comments

• Is 7900 xt basically idential to M2 ultra chips in terms of token generation speed?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 2 comments

• Model size increase in ollama when context size increases

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 16 comments

• What is your acceptable TG speed?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 50 comments

• New year (2025) poll: At what MSRP would you purchase RTX 5090 32GB?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 37 comments

• What is your acceptable PP speed?

  Posted by siegevjorn@reddit | LocalLLaMA | View on Reddit | 9 comments