Instructions to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF",
	filename="ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF
# Run inference directly in the terminal:
llama cli -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF
# Run inference directly in the terminal:
llama cli -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF
# Run inference directly in the terminal:
./llama-cli -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Use Docker

docker model run hf.co/maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

LM Studio
Jan
Ollama
How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with Ollama:
```
ollama run hf.co/maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF
```

Unsloth Studio

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF to start chatting

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with Docker Model Runner:
```
docker model run hf.co/maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF
```

Lemonade

How to use maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull maczzzzzz/ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF

Run and chat with the model

lemonade run user.ThinkingCap-Qwen3.6-27B-TQ3_4S-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

maczzzzzz commited on 4 days ago

Commit

9403c1d

verified ·

1 Parent(s): 386abba

Upload ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf + AEON bench data

Browse files

Files changed (4) hide show

.gitattributes +1 -0
BENCH-aeon-75-case-6144.md +42 -0
README.md +134 -0
ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf +3 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf filter=lfs diff=lfs merge=lfs -text

BENCH-aeon-75-case-6144.md ADDED Viewed

	@@ -0,0 +1,42 @@

+# AEON Bench — ThinkingCap-Qwen3.6-27B-TQ3_4S (75 cases @6144)
+**Date:** 2026-07-09 20:09 EDT
+**Framework:** AEON-Bench-Pod v1 (self-reported mode)
+**Hardware:** NVIDIA RTX 5060 Ti 16 GB (turbo-tan/llama.cpp-tq3 77dd77473, TQ3_4S)
+**Budget:** `--max-tokens 6144` blanket
+**Run shape:** `--limit 75 --fast` (no agentic harness)
+## Results
+**Overall mean: 0.480** across 75 scored / 75 cases
+| Category | Mean | N |
+|---|---|---|
+| math | 0.400 | 30 |
+| instruction | 0.300 | 30 |
+| reasoning | 1.000 | 15 |
+| Difficulty | Mean | N |
+|---|---|---|
+| easy | 1.000 | 6 |
+| medium | 0.778 | 9 |
+| hard | 0.733 | 15 |
+| expert | 0.429 | 21 |
+| frontier | 0.125 | 24 |
+## Comparison vs sibling Tess-4-27B-TQ3_4S (same harness)
+| Model | math | instruction | reasoning | mean |
+|---|---|---|---|---|
+| thinkingcap-27b-tq3-4s.gguf (this) | 0.400 | 0.300 | 1.000 | 0.480 |
+| tess-4-27b-tq3-4s.gguf | 0.500 | 0.333 | 0.933 | 0.560 |
+**ThinkingCap vs Tess profile (both 27B-TQ3_4S on node-d):**
+- Both struggle on instruction (ThinkingCap 0.300, Tess 0.333)
+- Both dominate reasoning — ThinkingCap scores a perfect 1.000, Tess 0.933
+- Math is the differentiator: Tess wins at 0.500 vs ThinkingCap 0.400
+## Notes
+- 30 prose cases (not in this 75-case run) require a frontier judge endpoint — excluded.
+- Bench run in self-reported mode. Mothership returned HTTP 403 (expected).

README.md ADDED Viewed

	@@ -0,0 +1,134 @@

+---
+license: apache-2.0
+base_model: bottlecapai/ThinkingCap-Qwen3.6-27B
+tags:
+- gguf
+- tq3
+- qwen35
+- quantization
+- llama-cpp
+- blackwell
+- cuda
+base_model_relation: quantized
+quantized_by: maczzzzzz (via turbo-tan/llama.cpp-tq3)
+---
+# ThinkingCap-Qwen3.6-27B TQ3_4S — GGUF
+**TQ3_4S quant of bottlecapai/ThinkingCap-Qwen3.6-27B (Apache 2.0), produced via turbo-tan's llama.cpp-tq3 CUDA fork.** ThinkingCap is a Qwen3.6-27B fine-tune optimized for reasoning-heavy tasks. Benchmarked on a Blackwell RTX 5060 Ti (16 GB). Mean AEON score: **0.480** (75 cases, 6144 budget).
+## File
+| File | Size | Quant | BPW |
+|---|---|---|---|
+| `ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf` | 14 GB | TQ3_4S (turbo-tan) | ~3.8 bpw |
+## NOT a stock llama.cpp quant
+TQ3_4S (TurboQuant 3-bit 4-state) is a custom weight format unique to [turbo-tan/llama.cpp-tq3](https://github.com/turbo-tan/llama.cpp-tq3). Stock llama.cpp and nixpkgs `llama-cpp` will exit with `unknown quantization` at load time. Use the `llama-server`/`llama-cli` from the tq3 fork.
+## Scope of these benchmarks — read this first
+**These numbers are a light baseline, not a thorough TQ3 evaluation.** The mesh's bench framework is built for production agent workload regression-detection on the local stack, not for the kind of multi-axis sweep that upstream quant maintainers typically publish. Specifically:
+- **Harness scope is bounded.** The numbers below come from the mesh's Aeon-Bench-Pod (self-reported mode, 75-case `--fast` shape, text-only, no agentic harness). That's a regression suite, not a quality benchmark.
+- **Sample sizes are small.** 75 cases on a single GPU, single rep. None are powered for statistical significance.
+- **No perplexity / wikitext / MMLU / GSM8K.** The mesh's stack isn't a quality benchmark — those are upstream's territory.
+- **Single GPU class (Blackwell 16 GB).** All measurements are on an NVIDIA RTX 5060 Ti 16 GB (CUDA 13.2, turbo-tan/llama.cpp-tq3 77dd77473). No RDNA4, no multi-GPU, no Vulkan. Cross-hardware generalization is NOT implied.
+- **No human eval.** "0.480 mean on the AEON suite" is not a quality verdict on this specific quant.
+- **30 prose cases unscored** — no frontier judge endpoint configured, so the prose category is excluded from the mean.
+**What this IS good for:** a quick signal that the quant (a) loads, (b) runs at sane throughput, (c) doesn't break the mesh's agent tool-calling. **What this is NOT good for:** claiming "this is the best quant of this model," reproducing academic benchmark results, or substituting for upstream's validation work.
+For a rigorous view, see [bottlecapai/ThinkingCap-Qwen3.6-27B](https://huggingface.co/bottlecapai/ThinkingCap-Qwen3.6-27B) (parent model) and [turbo-tan/llama.cpp-tq3](https://github.com/turbo-tan/llama.cpp-tq3) (quantizer). Raw bench reports are attached as `BENCH-*.md` files in this repo.
+## What we measured
+### AEON v3 — 75 cases (self-reported, 6144 budget)
+Benchmarked on Blackwell RTX 5060 Ti 16 GB, `--limit 75 --fast --max-tokens 6144`.
+| Category | Mean | N |
+|---|---|---|
+| **Overall** | **0.480** | 75 |
+| math | 0.400 | 30 |
+| instruction | 0.300 | 30 |
+| reasoning | 1.000 | 15 |
+| Difficulty | Mean | N |
+|---|---|---|
+| easy | 1.000 | 6 |
+| medium | 0.778 | 9 |
+| hard | 0.733 | 15 |
+| expert | 0.429 | 21 |
+| frontier | 0.125 | 24 |
+**ThinkingCap's profile:** pure reasoning specialist (1.000 — perfect on reasoning). Weak on instruction (0.300) and math (0.400). Best suited for roles where reasoning depth is the priority.
+### Comparison vs. sibling quants (75-case @6144, same harness, node-d)
+| Model | mean | math | instruction | reasoning |
+|---|---|---|---|---|
+| Tess-4-27B-TQ3_4S | 0.560 | 0.500 | 0.333 | 0.933 |
+| **ThinkingCap-Qwen3.6-27B-TQ3_4S** | **0.480** | 0.400 | 0.300 | 1.000 |
+| Qwen3.6-27B-TQ3_4S (prod baseline) | 0.520 | 0.500 | 0.333 | 0.933 |
+## Quick start
+```bash
+# Build turbo-tan's tq3 fork
+git clone https://github.com/turbo-tan/llama.cpp-tq3
+cd llama.cpp-tq3
+mkdir build && cd build
+cmake .. -DGGML_CUDA=ON -DCUDA_DOCKER_ARCH=sm_120
+make -j$(nproc)
+# Serve
+llama-server \
+  -m ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf \
+  --host 0.0.0.0 --port 8081 \
+  -ngl 99 -c 131072 -t 12 \
+  -ctk q4_0 -ctv q4_0 \
+  -np 1 --batch-size 512 --ubatch-size 128 \
+  --jinja --metrics -rea off
+```
+## Reproduce the quant
+```bash
+# Requires the tq3 fork and the F16 source GGUF
+llama-quantize --allow-requantize ThinkingCap-Qwen3.6-27B-f16.gguf \
+  ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf TQ3_4S
+```
+## Files in this repo
+| File | Description |
+|---|---|
+| `ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf` | The quantized model (LFS-tracked) |
+| `README.md` | This model card |
+| `BENCH-aeon-75-case-6144.md` | AEON bench results (75 cases, 6144 budget) |
+## What's NOT in this repo (caveats)
+- **Stock llama.cpp will not load this file.** TQ3_4S is a custom weight format unique to turbo-tan/llama.cpp-tq3.
+- **No AMD GPU bench.** All measurements are RTX 5060 Ti (CUDA). No RDNA4, no ROCm.
+- **No quality benchmark** (perplexity, MMLU, GSM8K). The custom 3-bit quant works on the mesh's regression tests; whether it's "the best TQ3 quant" needs upstream validation.
+- **No MTP / speculative-decode bench.** ThinkingCap was tested without MTP.
+- **No prose/creativity evaluation.** 30 prose cases were unscored (no judge endpoint configured).
+## Provenance
+- **Source model:** [bottlecapai/ThinkingCap-Qwen3.6-27B](https://huggingface.co/bottlecapai/ThinkingCap-Qwen3.6-27B) — Qwen3.6-27B fine-tune
+- **Source model license:** Apache 2.0
+- **Quantizer:** [turbo-tan/llama.cpp-tq3](https://github.com/turbo-tan/llama.cpp-tq3) @ 77dd77473
+- **Quantizer license:** MIT
+- **Build hardware:** NVIDIA RTX 5060 Ti 16 GB (Blackwell), CUDA 13.2, NixOS 25.11
+- **Bench harness:** Aeon-Bench-Pod v1 (self-reported mode)
+- **Original bench report:** `raw/benchmarks/2026-07-09-node-d-thinkingcap-tq3/` in the meshina repo
+## License
+- **ThinkingCap-Qwen3.6-27B** is Apache 2.0 (per its HF model card).
+- **turbo-tan/llama.cpp-tq3** is MIT.
+- The GGUF in this repo is a derivative of the Apache 2.0-licensed parent, produced with the MIT-licensed quantizer. The Apache 2.0 license is preserved.

ThinkingCap-Qwen3.6-27B-TQ3_4S.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:efe129e3902be20d2c06943ae7f5995a2d17119012e796bcb77f5b73f6382f65
+size 14167400416