Instructions to use Verdugie/STEM-Oracle with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Verdugie/STEM-Oracle 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 Verdugie/STEM-Oracle:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Verdugie/STEM-Oracle:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Verdugie/STEM-Oracle:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Verdugie/STEM-Oracle:Q4_K_M

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 Verdugie/STEM-Oracle:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Verdugie/STEM-Oracle:Q4_K_M

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 Verdugie/STEM-Oracle:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Verdugie/STEM-Oracle:Q4_K_M

Use Docker

docker model run hf.co/Verdugie/STEM-Oracle:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use Verdugie/STEM-Oracle with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Verdugie/STEM-Oracle"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Verdugie/STEM-Oracle",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Verdugie/STEM-Oracle:Q4_K_M

Ollama
How to use Verdugie/STEM-Oracle with Ollama:
```
ollama run hf.co/Verdugie/STEM-Oracle:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use Verdugie/STEM-Oracle with Docker Model Runner:
```
docker model run hf.co/Verdugie/STEM-Oracle:Q4_K_M
```

Lemonade

How to use Verdugie/STEM-Oracle with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Verdugie/STEM-Oracle:Q4_K_M

Run and chat with the model

lemonade run user.STEM-Oracle-Q4_K_M

List all available models

lemonade list

Atomic Chat

Verdugie commited on Mar 11

Commit

62092f7

verified ·

1 Parent(s): 509bd5c

Initial model card — stress testing in progress

Browse files

Files changed (1) hide show

README.md +176 -0

README.md ADDED Viewed

	@@ -0,0 +1,176 @@

+---
+license: apache-2.0
+language:
+  - en
+  - es
+base_model: Qwen/Qwen3.5-27B
+tags:
+  - conversational
+  - stem
+  - tutor
+  - math
+  - physics
+  - chemistry
+  - biology
+  - computer-science
+  - gguf
+  - claude-distillation
+  - opus
+  - density-optimized
+  - 27b-dense
+library_name: transformers
+pipeline_tag: text-generation
+---
+> **Stress testing in progress.** Results and selected transcripts will be added once the full test battery completes across all quants.
+# STEM-Oracle-27B
+A STEM tutor that doesn't hold your hand — it holds you accountable. Fine-tuned from **Qwen 3.5 27B Dense** on **5,179 conversations** distilled from Claude Opus 4.6, purpose-built for teaching science, mathematics, physics, chemistry, biology, and computer science.
+STEM-Oracle shares the oracle-soul architecture with [Opus-Candid-27B-V3.5](https://huggingface.co/Verdugie/Opus-Candid-27B-V3.5) — same 6-dimensional Zipf scoring, same parameter-aware density equilibrium, same quantization survival strategy — but the training data is entirely different. Where V3.5 trains on personality and adversarial resistance, STEM-Oracle trains on tiered STEM pedagogy, error correction, Socratic method, and cross-domain bridges.
+---
+## What Makes This Different from Math-Distilled Models
+Standard STEM fine-tunes (WizardMath, MetaMath, etc.) train on solution-answer pairs. The model learns to pattern-match problem structures to solution templates. Works on benchmarks. Breaks on follow-up questions.
+STEM-Oracle trains the *reasoning disposition* alongside the domain knowledge:
+- **Tiered depth** — the same concept explained at five levels, from freshman intuition to graduate formalism. The model meets you where you are, not where it wants to be.
+- **Error correction without condescension** — catches misconceptions and wrong steps, explains *why* they're wrong, not just *that* they're wrong.
+- **Socratic method** — asks probing questions instead of immediately giving answers. Forces understanding over memorization.
+- **Cross-domain bridges** — connects linear algebra to quantum mechanics, graph theory to chemistry, thermodynamics to information theory. Trained to make connections conventional tutors don't.
+- **Sustained coherence** — holds context across 10+ turn problem-solving sessions without contradicting earlier steps or losing the thread.
+---
+## Available Quantizations
+| File | Quant | Size | Notes |
+|------|-------|------|-------|
+| `STEM-Oracle-27B-Q8_0.gguf` | Q8_0 | ~28 GB | Reference quality. |
+Additional quantizations (Q4_K_M, Q6_K) will be added after stress testing validates personality and accuracy survival across quant levels.
+---
+## Model Details
+| Attribute | Value |
+|-----------|-------|
+| **Base Model** | Qwen 3.5 27B Dense (hybrid Mamba-Transformer) |
+| **Training Data** | 5,179 STEM-focused multi-turn conversations with Claude Opus 4.6 |
+| **Dataset Architecture** | 6-dimensional Zipf scoring + parameter-aware density equilibrium |
+| **Fine-tune Method** | LoRA + rsLoRA (r=128, alpha=256) via PEFT + TRL |
+| **Training Hardware** | NVIDIA A100 SXM 80GB (RunPod) |
+| **Precision** | bf16 |
+| **Optimizer** | AdamW 8-bit |
+| **Learning Rate** | 5e-5 (cosine schedule, 6% warmup — tuned for SSM stability) |
+| **License** | Apache 2.0 |
+---
+## Quick Start
+Works with any GGUF-compatible runtime — LM Studio, Ollama, llama.cpp, KoboldCpp. Download the GGUF, load it, and start asking questions. No system prompt needed — the teaching disposition is in the weights.
+---
+## Recommended Hardware
+| Setup | Quantization | VRAM/RAM | Speed | Notes |
+|-------|-------------|----------|-------|-------|
+| **RTX 4090 (24GB)** | Q8_0 | ~28 GB VRAM | 8-15 t/s | Tight fit, may need partial offload. |
+| **Apple M2/M3 Ultra** | Q8_0 | 64-128 GB unified | 5-10 t/s | Full model in unified memory. |
+| **Dual GPU** | Q8_0 | ~30 GB VRAM | Varies | Split across two 16GB+ cards. |
+---
+## Test Battery Design
+STEM-Oracle's stress test protocol covers 40 single-turn prompts and 10 multi-turn problem-solving conversations:
+**Single-turn (40 prompts):**
+- Identity (6) — knows what it is and what it isn't
+- Math (6) — tiered from basic calculus through functional analysis
+- Physics (5) — Newton through gauge invariance
+- Chemistry (4) — bonds through NMR spectroscopy
+- Biology (3) — natural selection through CRISPR mechanisms
+- Computer Science (3) — Big-O through the halting problem
+- Error Correction (6) — catches student misconceptions
+- Cross-Domain Bridges (3) — connects disciplines
+- Conciseness (3) — quick factual density checks
+**Multi-turn (10 conversations, 70+ turns):**
+- Derivative deep dives with struggling freshmen
+- Physics problem-solving with wrong intermediate steps
+- Organic chemistry mechanisms with tier-shifting
+- Proof guidance without giving the answer
+- Socratic questioning that builds understanding
+- Extended adversarial challenges to mathematical claims
+Results pending.
+---
+## The Oracle-Soul Architecture
+STEM-Oracle and [Opus-Candid-27B-V3.5](https://huggingface.co/Verdugie/Opus-Candid-27B-V3.5) share the same training architecture — the difference is what fills it:
+| Dimension | V3.5 (Personality) | STEM-Oracle (Tutor) |
+|-----------|-------------------|---------------------|
+| Training conversations | 5,358 | 5,179 |
+| Primary signal | Personality, adversarial resistance | STEM pedagogy, error correction |
+| Density equilibrium | 36-40w median | Adaptive — concise for facts, extended for derivations |
+| Reinforcement nodes | Identity, worth, trust, vulnerability, control | Accuracy, tier-matching, Socratic method, patience, bridges |
+| Anti-pattern training | Anti-sycophancy, anti-therapy-speak | Anti-hand-holding, anti-pattern-matching |
+Both models prove the same thesis: personality (or pedagogical disposition) can be trained into weights at a level that survives quantization, rather than bolted on via system prompts that any user can override.
+---
+## Choosing Your Model
+| Model | Best For | VRAM |
+|-------|----------|------|
+| [Lite 4B](https://huggingface.co/Verdugie/Opus-Candid-Lite-4B) | Phones, Raspberry Pi, integrated graphics | ~3 GB |
+| [8B V3](https://huggingface.co/Verdugie/Opus-Candid-8B-V3) | Fast casual chat, anything with 8GB VRAM | ~8 GB |
+| [MoE V3](https://huggingface.co/Verdugie/Opus-Candid-MoE-V3) | Best depth-per-VRAM ratio | ~22 GB |
+| [27B V3](https://huggingface.co/Verdugie/Opus-Candid-27B-V3) | Full experience, dense reasoning | ~27 GB |
+| [27B V3.5](https://huggingface.co/Verdugie/Opus-Candid-27B-V3.5) | Maximum personality depth | ~18-27 GB |
+| **STEM-Oracle-27B** (this model) | STEM tutoring, problem-solving, teaching | ~28 GB |
+The Opus Candid models are built for personality and conversation. STEM-Oracle is built for teaching. If you want a model that pushes back on bad arguments, run V3.5. If you want a model that catches your algebra mistakes and walks you through the fix, run this.
+---
+## Opus Candid Model Family
+| Model | Size | Base | Status |
+|-------|------|------|--------|
+| [Opus-Candid-Lite-4B](https://huggingface.co/Verdugie/Opus-Candid-Lite-4B) | 4B | Qwen 3 4B | Active |
+| [Opus-Candid-8B-V3](https://huggingface.co/Verdugie/Opus-Candid-8B-V3) | 8B | Qwen 3 8B | Active |
+| [Opus-Candid-MoE-V3](https://huggingface.co/Verdugie/Opus-Candid-MoE-V3) | 31B/3B | Qwen 3 30B-A3B | Active |
+| [Opus-Candid-27B-V3](https://huggingface.co/Verdugie/Opus-Candid-27B-V3) | 27B | Qwen 3.5 27B | Active |
+| [Opus-Candid-27B-V3.5](https://huggingface.co/Verdugie/Opus-Candid-27B-V3.5) | 27B | Qwen 3.5 27B | Active |
+| **STEM-Oracle-27B** (this model) | 27B | Qwen 3.5 27B | Active — Stress testing in progress |
+| [Opus-Candid-70B-V1](https://huggingface.co/Verdugie/Opus-Candid-70B-V1) | 72B | Qwen 2.5 72B | Archived |
+| [Opus-Candid-32B-V1](https://huggingface.co/Verdugie/Opus-Candid-32B-V1) | 32B | Qwen 2.5 32B | Archived |
+| [Opus-Candid-14B-V1](https://huggingface.co/Verdugie/Opus-Candid-14B-V1) | 14B | Qwen 2.5 14B | Archived |
+| [Opus-Candid-8B-V1](https://huggingface.co/Verdugie/Opus-Candid-8B-V1) | 8B | Qwen 2.5 7B | Archived |
+| [Opus-Research-8B-V1.5](https://huggingface.co/Verdugie/Opus-Research-8B-V1.5) | 8B | Qwen 2.5 7B | Archived |
+---
+## Dataset
+Training data will be available at [Verdugie/opus-candid-training-data](https://huggingface.co/datasets/Verdugie/opus-candid-training-data). ShareGPT format, Apache 2.0, compatible with TRL, Axolotl, and LLaMA-Factory.
+**License:** Apache 2.0. Open weight. No guardrails.
+---
+*Built by [Saul Verdugo](https://huggingface.co/Verdugie) — independent ML researcher. OpusReasoning@proton.me*
+<!-- build: 2026-03-10 12:45 -->