GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

brosnanyuen/Qwen3.6-27B-LTSpice-v2-GGUF overview

Experimental Qwen3.6 27B Finetune for creating LTSpice Circuit .asc files and simulating them Generate LTSpice Circuits using LLMs ./qwen.gif Generate LTSpice …

transformersggufunslothqwenqwen3_5image-text-to-textbase_model:Qwen/Qwen3.6-27Bbase_model:quantized:Qwen/Qwen3.6-27Blicense:apache-2.0endpoints_compatibleregion:usimatrixconversational

Runs locally from ~884.6 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
330
Likes
2
Pipeline
image-text-to-text

Repository Files & Downloads

26 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
BF16/Qwen3.6-27B-BF16-00001-of-00002.ggufGGUFBF1646.57 GBDownload
BF16/Qwen3.6-27B-BF16-00002-of-00002.ggufGGUFBF163.54 GBDownload
Qwen3.6-27B-IQ4_NL.ggufGGUFIQ4_NL14.97 GBDownload
Qwen3.6-27B-IQ4_XS.ggufGGUFIQ4_XS14.38 GBDownload
Qwen3.6-27B-Q3_K_M.ggufGGUFQ3_K_M12.65 GBDownload
Qwen3.6-27B-Q3_K_S.ggufGGUFQ3_K_S11.51 GBDownload
Qwen3.6-27B-Q4_0.ggufGGUFQ4_014.71 GBDownload
Qwen3.6-27B-Q4_1.ggufGGUFQ4_116.07 GBDownload
Qwen3.6-27B-Q4_K_M.ggufGGUFQ4_K_M15.66 GBDownload
Qwen3.6-27B-Q4_K_S.ggufGGUFQ4_K_S14.77 GBDownload
Qwen3.6-27B-Q5_K_M.ggufGGUFQ5_K_M18.17 GBDownload
Qwen3.6-27B-Q5_K_S.ggufGGUFQ5_K_S17.66 GBDownload
Qwen3.6-27B-Q6_K.ggufGGUFQ6_K20.98 GBDownload
Qwen3.6-27B-Q8_0.ggufGGUFQ8_026.63 GBDownload
Qwen3.6-27B-UD-IQ2_M.ggufGGUFIQ2_M10.10 GBDownload
Qwen3.6-27B-UD-IQ2_XXS.ggufGGUFIQ2_XXS8.74 GBDownload
Qwen3.6-27B-UD-IQ3_XXS.ggufGGUFIQ3_XXS11.17 GBDownload
Qwen3.6-27B-UD-Q2_K_XL.ggufGGUFQ2_K_XL11.04 GBDownload
Qwen3.6-27B-UD-Q3_K_XL.ggufGGUFQ3_K_XL13.48 GBDownload
Qwen3.6-27B-UD-Q4_K_XL.ggufGGUFQ4_K_XL16.40 GBDownload
Qwen3.6-27B-UD-Q5_K_XL.ggufGGUFQ5_K_XL18.66 GBDownload
Qwen3.6-27B-UD-Q6_K_XL.ggufGGUFQ6_K_XL23.88 GBDownload
Qwen3.6-27B-UD-Q8_K_XL.ggufGGUFQ8_K_XL32.90 GBDownload
mmproj-BF16.ggufGGUFBF16888.0 MBDownload
mmproj-F16.ggufGGUFF16884.6 MBDownload
mmproj-F32.ggufGGUFF321.72 GBDownload

Model Details

Model IDbrosnanyuen/Qwen3.6-27B-LTSpice-v2-GGUF
Authorbrosnanyuen
Pipelineimage-text-to-text
Licenseapache-2.0
Base modelQwen/Qwen3.6-27B
Last modified2026-07-24T03:01:38.000Z

Model README

---

library_name: transformers

license: apache-2.0

license_link: https://huggingface.co/Qwen/Qwen3.6-27B/blob/main/LICENSE

pipeline_tag: image-text-to-text

base_model:

  • Qwen/Qwen3.6-27B

tags:

  • unsloth
  • qwen
  • qwen3_5

---

Experimental Qwen3.6 27B Finetune for creating LTSpice Circuit .asc files and simulating them

!Generate LTSpice Circuits using LLMs

!Generate LTSpice Circuits using LLMs

!Converts ltspice netlist .net to ltspice .asc file format!

!REALTIME LTSpice simulation and export to .csv!

https://github.com/BrosnanYuen/LLM_Circuit_Design_Using_LTSpice_Tutorial/

Install

Install these first and add to your harness

  • https://github.com/BrosnanYuen/symbolic_math_mcp
  • https://github.com/BrosnanYuen/bltspice_mcp

Example Workflow Prompt

You are an electrical engineering assistant specializing in circuit design and LTSpice

Goal: Design an active bandpass filter using opamps that has cutoff frequency between 30 Hz and 25 KHz

where the cutoff frequency has half the power of the peak. Use +12V and -12V rails.

There are four phases for the the design, ONLY FOLLOW IN THIS ORDER:

  • Phase 1: Search the web and examples for the optimal design
  • Phase 2: Calculate the values for the components and verify the calculations using symbolic_math_mcp
  • Phase 3: Use the calculations to create a LTSpice netlist file .net and simulate to verify the results using ltspice_mcp
  • Phase 4: Finally, convert LTSpice netlist file .net to LTSpice asc file .asc using ltspice_mcp

If something is wrong delete everything and start from begining

ONLY WRITE TO FILES INSIDE THIS FOLDER

DO NOT WRITE TO ANY FILES OUTSIDE THIS FOLDER

CAN READ FILES OUTSIDE THIS FOLDER

Phase 1: Research circuit design

Search the web for circuit designs that can be created and simulated in LTSpice

Read some .yaml and .net files in ./examples/

Phase 2: Calculate values for circuit design

1) Read ./YAML_tutorial.md and some of the example .yaml files in ./examples/ to create a new ./circuit.yaml containing the calculations for the circuit design

Use simple equations and calculations.

2) Read ./symbolic_math_mcp_for_LLM.md on how to use the symbolic_math_mcp server to verify the newly created ./circuit.yaml

Wait atleast 10 mins for symbolic_math_mcp tool call to finish

Phase 3: Create the LTSpice netlist file .net

1) Read ./LTSPICE_NET.md and some of the example .net files in ./examples/ to create a new LTSpice netlist ./circuit.net using the calculations above

DO NOT USE ANY COMPONENTS OUTSIDE OF THE LTSPICE library of ~/.wine/drive_c/users/brosnan/AppData/Local/LTspice/

2) Read ./bltspice_mcp_for_LLM.md and ./run_ltspice_netlist_to_csv.md to simulate the LTSpice netlist ./circuit.net and get a .csv file of simulation to verify the circuit is designed correctly

3) Write python code to read the .csv files and verify the circuit design works.

MUST HAVE: 30 Hz and 25 KHz cutoff freq

Phase 4: Convert the LTSpice .net to .asc

Read ./bltspice_mcp_for_LLM.md and use ltspice_netlist_to_asc from ltspice_mcp to convert LTSpice .net files to LTSpice .asc files

ONLY READ ./LTSPICE_ERROR_CODES.md if have errors

Run brosnanyuen/Qwen3.6-27B-LTSpice-v2-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models