Page Title: Ai - Lm Studio Models
Last Modified: 2026-11-06 14:03:53
Current Date: 2026-11-06 14:18:44 UTC

LM Studio is a desktop app that lets you run AI language models on your own computer instead of sending everything to the cloud.

I use a couple of models on LM Studio depending on what I am doing.

Nerdsking
  • Nerdsking-python-coder-7B-i_Q5_K_M.gguf is a GGUF file for the Nerdsking Python coder model, stored in a Q5_K_M quantized format. That means it is compressed to use less memory and run faster than full-precision weights, while keeping decent quality. It is best for local Python coding help, function writing, debugging, and lightweight developer workflows on consumer hardware
  • Model Nerdsking/Nerdsking-python-coder-7B-i
  • File nerdsking-python-coder-7B-i_Q5_k_m.gguf
  • Format GGUF
  • Quantization Q5_K_M
  • Arch qwen2
  • Capabilities Tool use
  • Domain 11m
  • Size on disk 5.44GB
Qwen 3.6
  • Qwen3.6-35B-A3B is an open-weight multimodal model with 35 billion total parameters and about 3 billion active per token. It’s built for efficient inference, long-context work, tool use, and “thinking”/reasoning across multi-turn chats. It’s best for coding agents, repository-scale code review, frontend and terminal workflows, structured outputs, and multimodal tasks using text, images, and video
  • Model qwen/qwen3.6-35b-a3b
  • File Qwen3.6-35B-A3B-Q4_K_M.gguf
  • Format GGUF
  • Quantization Q4_K_M
  • Arch qwen35moe
  • Capabilities Vision, Tool use & Reasoning
  • Domain 11m
  • Size on disk 22.07GB
Qwen3 Coder
  • Qwen3-Coder-30B-A3B-Instruct is a 30.5B-total, 3.3B-active Mixture-of-Experts coding model built for agentic coding, tool use, browser use, and repository-scale context. It has native 262K-token context and is designed for efficient, high-throughput code work. It’s best for coding agents, test generation, code review, large-codebase understanding, and developer tools that need strong performance at lower serving cost than bigger flagship models
  • Model qwen/qwen3-coder-30b
  • File Qwen3-Coder-30B-A38-Instruct-Q4_K_M.gguf
  • Format GGUF
  • Quantization Q4_K_M
  • Arch qwen35moe
  • Capabilities Tool use
  • Domain 11m
  • Size on disk 18.63GB
Gemma4
  • Gemma 4 is a family of open AI models from Google designed for strong reasoning, multimodal understanding, and agentic workflows. It comes in smaller on-device versions and larger models for PCs and servers, so it can run locally or at scale. It’s best for building chatbots, coding assistants, summarization tools, multilingual apps, and autonomous agents that can plan, call functions, and handle text, audio, and vision
  • Model google/gemma-4-a4b-qat
  • File gemma-4-26B-A4B-it-QAT-Q4_0.gguf
  • Format GGUF
  • Quantization Q4_0
  • Arch gemma4
  • Capabilities Vision, Tool use & Reasoning
  • Domain 11m
  • Size on disk 15.63GB

Copyright © 2002 to 2026