TGTGInsighttelegram intelligenceLIVE / telegram public index
← GitHub Trends

TGINSIGHT SIMILAR POSTS

Find similar content

Source channel @githubtrending · Post #14974 · Jul 19

#cplusplus ik_llama.cpp is an improved version of llama.cpp that runs faster on CPUs and hybrid GPU/CPU setups. It supports many new advanced quantization methods, which help models use less memory and run more efficiently. It also offers better performance for special models like DeepSeek and MoE, with faster prompt processing and token generation. You can run it on various hardware, including Android, and it has features to control where model data is stored (CPU or GPU). This means you get quicker AI responses and can handle bigger or more complex models smoothly on your computer or device[2][1][4]. https://github.com/ikawrakow/ik_llama.cpp

Hashtags

Results

1 similar post found

Search: #localllama

当前筛选 #localllama清除筛选
是芙莉莲

@ireallyhatetheworld · Post #1459 · 03/16/2026, 01:23 PM

Qwen3.5-9B-Claude-4.6-Opus-Uncensored-Distilled-GGUF: 面向本地部署的轻量级创意与推理模型 🔞可用于本地涩涩等场景 • 基于 Qwen 3.5 9B,并融入 Claude Opus 4.6 蒸馏思路,主打更强的创意表达、对话表现与角色扮演场景 • 提供 GGUF 与低显存友好的 Q4_K_M 量化版本,作者反馈在 RTX 3060 12 GB 上可达约 38 tok/s,适合本地聊天、游戏 NPC 与 Home Lab 部署 • 默认关闭 thinking 以提升通用聊天体验,需要时可在 LM Studio 中手动开启;模型采用 Apache 2.0 许可证,便于社区测试与二次集成 https://www.reddit.com/r/LocalLLaMA/comments/1runlpf/qwen359bclaude46opusuncensoreddistilledgguf #AI#Uncensored#本地大模型#模型蒸馏#GGUF#Qwen#Claude#LMStudio#量化模型#低显存部署#角色扮演#LocalLLaMA