TGTGInsighttelegram intelligenceLIVE / telegram public index
← python-telegram-bot

TGINSIGHT SIMILAR POSTS

Find similar content

Source channel @pythontelegrambotchannel · Post #89 · Oct 7

The v13 release is not just a release either, it is also our official announcement of participation in the annual #hacktoberfest. 💻🥨 We know that we're a few days late to the party, but v13 had to get ready before. 😉 This year, the fest is opt-in for projects and we definitely want to opt into taking part in this great event! If you ever thought about starting coding or giving back to your favourite open source repositories, now is the time! Head over to the hacktoberfest website to learn more about it. We already prepared some issues on our repositories and aim towards opening more issues for starters, but feel free to begin a hunt for improvements and fixes by yourself!

Results

3 similar posts found

Search: #lmstudio

当前筛选 #lmstudio清除筛选
Machinelearning

@ai_machinelearning_big_data · Post #9311 · 12/24/2025, 10:45 AM

📌Тренируем FunctionGemma самостоятельно. LM Studio в коллаборации с Unsloth опубликовали подробный туториал по файнтюнингу недавно выпущенной Google модели FunctionGemma. FunctionGemma - уменьшенная версия Gemma (всего 270Ь параметров) для агентских сценариев и работы в качестве бэкенда приложений, которую можно запускать практически на любом устройстве. Гайд состоит из подробного описания всего процесса от обучения модели вызову инструментов до преобразования в GGUF-формат и последующего запуска его в LM Studio Туториал подойдет для локального трейна (Unsloth работает на NVIDIA, AMD и Intel), но есть и готовый Collab Notebook для тренировки в облаке. ⚠️ FunctionGemma не предназначена для использования в качестве прямой диалоговой модели. @ai_machinelearning_big_data #AI#ML#LLM#Tutorial#Unsloth#LMStudio

Crypto M - Crypto News

@CryptoM · Post #65303 · 04/12/2026, 05:19 PM

🚀 AI TRENDS | New Local Model Qwopus3.5-27B-v3 Released with High HumanEval Score Developer Jackrong has introduced Qwopus3.5-27B-v3, a local model designed to operate on a single consumer GPU. According to NS3.AI, this model boasts an impressive 95.73% score on HumanEval. The Qwopus3.5-27B-v3 is distilled from Claude Opus 4.6-style reasoning and is available in GGUF format for use with LM Studio or llama.cpp. #AI#Qwopus3.5 #HumanEval#Jackrong#ClaudeOpus#GGUF#LMStudio#llama_cpp#AITrends

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