Downloads · 30 days
232
28% of all-time downloads
sigmanih/sigma-alpaca-3b-gguf
sigma-alpaca-3b-gguf is a text generation model from sigmanih. Use it when you need the model to write or continue text. The card lists the license as other.
Downloads · 30 days
232
28% of all-time downloads
All-time downloads
834
Public
Repo size
2 GB
Likes
1
Public
Click a slice to open those files.
.gguf2 GB · 100%
From the Hugging Face model README
❤️ Support & Community: If you find this model helpful, please give this repository a Like on Hugging Face and a ⭐ Star on our SigmaStudio GitHub!
sigma-alpaca-3b-gguf is a production-ready model optimized and published using the Model Hub module of Sigma Studio.
| Specification | Value |
|---|---|
| Model Repository | sigmanih/sigma-alpaca-3b-gguf |
| Weight Format | Sconosciuto (sconosciuta) |
| Base Architecture | CausalLM |
| Active Parameters | 3B |
| Context Window | 32,768 tokens |
| Total Disk Footprint | 0.0 GB |
| Inference RAM / VRAM | ~3.2 GB VRAM (Full GPU offload) / ~2.9 GB RAM (CPU/Hybrid) |
| Recommended Hardware | GPU with 6-8 GB VRAM (or 16 GB system RAM) |
| Recommended Usage | Edge devices, Real-time voice agents, Mobile & CPU-friendly workloads. |
Evaluated directly on GPU via Sigma Studio Training Lab (Deterministic seed 42, Temp 0.0):
| Benchmark Suite | Score / Accuracy | Total Questions Evaluated | Pass Rate | Test Date | Execution Engine |
|---|---|---|---|---|---|
| Tutti i Benchmark Ufficiali | 42.0% | 42/100 quesiti superati | 42.0% Pass | 2026-08-30 | ⚡ SigmaEngine Direct GPU |
Side-by-side performance comparison between direct zero-overhead answer (No-Thinking) and Chain-of-Thought step-by-step reasoning (Deep Thinking CoT):
⚡ No-Thinking (Direct Response) : [████████░░░░░░░░░░░░] 42.1% (24/57 passed)
🧠 Deep Thinking (CoT Reasoning) : [████████░░░░░░░░░░░░] 41.9% (18/43 passed)
📈 CoT Performance Delta : -0.2%
%%{init: {'theme': 'dark'}}%%
xychart-beta
title "Accuracy (%): No-Thinking vs Deep Thinking"
x-axis ["⚡ No-Thinking (Direct)", "🧠 Deep Thinking (CoT)"]
y-axis "Accuracy (%)" 0 --> 100
bar [42.1, 41.9]
| Execution Mode | Accuracy (%) | Questions Passed | Operational Profile & Latency |
|---|---|---|---|
| ⚡ No-Thinking (Direct Response) | 42.1% | 24 / 57 | Minimal latency, immediate token-to-first-byte, zero reasoning tokens overhead |
| 🧠 Deep Thinking (CoT Reasoning) | 41.9% | 18 / 43 | Multi-step structured reasoning trace (-0.2%), optimal for math & hard logic |
| Dataset / Benchmark Suite | Domain / Category | Correct / Total | Accuracy (%) | Status |
|---|---|---|---|---|
| ARC-Challenge | Science & Grade-School Reasoning | 6 / 9 | 67% | ⚡ Fair |
| BIG-Bench Hard | Complex Multi-Task Logic & Symbolics | 2 / 7 | 29% | ⚠️ Low |
| GPQA | Graduate-Level Academic Reasoning | 4 / 9 | 44% | ⚡ Fair |
| GSM8K | Multi-Step Grade School Math | 6 / 9 | 67% | ⚡ Fair |
| HellaSwag | Commonsense Reasoning & Situational NLI | 5 / 9 | 56% | ⚡ Fair |
| HumanEval | Python Coding (pass@1) | 3 / 7 | 43% | ⚡ Fair |
| MATH | Championship Competition Math | 4 / 9 | 44% | ⚡ Fair |
| MBPP | Python Programming with Unit Tests | 4 / 9 | 44% | ⚡ Fair |
| MMLU | General Knowledge & Multi-Subject | 4 / 14 | 29% | ⚠️ Low |
| MMLU-Pro | Advanced Multi-Step Reasoning | 2 / 9 | 22% | ⚠️ Low |
| TruthfulQA | Factuality & Anti-Hallucination | 2 / 9 | 22% | ⚠️ Low |
| 🏆 OVERALL TOTAL | All Evaluated Datasets | 42 / 100 | 42% | 🏆 42% Pass |
Protocol: code_execution, continuation_logprob, cot_generation, letter_logprob · temp 0.0 · seed 42
Reproducibility hash: SHA256-C3F37AE9AE81ADCB
⚠️ Measured on a slice of the dataset, not the full suite: this score is not comparable with a full-suite run.
Empirical multi-turn agent reliability evaluation (zero quizzes, fully grounded file-system actions inside isolated Sandbox Jail):
| Protocol Metric | Outcome | Validation Criteria & Details |
|---|---|---|
| Tool Protocol Adherence | 60.9% | 0/92 analytical criteria verified |
| Autonomous Task Completion | 0.0% | 0/1 scenarios completed with exact target file |
| Sandbox Jail Containment | 100% Compliant | Zero escape attempts outside workspace boundary |
| Execution Efficiency | 0 turns (0.0s) | Optimal multi-step turn and token budget usage |
Measured on NVIDIA GeForce RTX 5070 Ti • 15.9 GB VRAM. Two different numbers follow, and they are not interchangeable.
| What was measured | Value | How |
|---|---|---|
| Single-stream decode (what a chat feels) | 200.2 tok/s | one request at a time, on NVIDIA GeForce RTX 5070 Ti |
| Prompt processing | 1904 tok/s | same probe |
| Aggregate throughput during evaluation | 192.1 tok/s | several requests in flight — not what a single answer runs at |
Speeds on other hardware were not measured and are not guessed here. A single-stream figure from one machine cannot be scaled into a prediction for another: it depends on memory bandwidth, quantization, context length and driver, and the error is large enough to be misleading.
Launch Sigma Studio to enjoy full 1-click GPU hardware acceleration, live monitoring, and visual chat:
# Clone and run Sigma Studio
git clone https://github.com/Sigmanih/SigmaStudio.git
cd SigmaStudio
.\sigma_studio.bat
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_id = "sigmanih/sigma-alpaca-3b-gguf"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto")
sigma-alpaca-3b-gguf è un modello ottimizzato pronto per l'inferenza e l'integrazione locale, pubblicato attraverso Σ-SIGMA Studio.
CausalLM (3B parametri)Sconosciuto (sconosciuta)0.0 GB~3.2 GB VRAM (offload GPU completo) | ~2.9 GB RAM (inferenza CPU/ibrida)32,768 tokenTutti i Benchmark Ufficiali42.0% (42/100 quesiti superati)Confronto visuale tra risposta istantanea diretta (No-Thinking) e ragionamento guidato multi-step (Deep Thinking CoT):
⚡ No-Thinking (Risposta Diretta) : [████████░░░░░░░░░░░░] 42.1% (24/57 superati)
🧠 Deep Thinking (CoT Reasoning) : [████████░░░░░░░░░░░░] 41.9% (18/43 superati)
📈 Delta Prestazionale CoT : -0.2%
%%{init: {'theme': 'dark'}}%%
xychart-beta
title "Accuratezza (%): No-Thinking vs Deep Thinking"
x-axis ["⚡ No-Thinking (Diretto)", "🧠 Deep Thinking (CoT)"]
y-axis "Accuratezza (%)" 0 --> 100
bar [42.1, 41.9]
| Modalità di Esecuzione | Accuratezza (%) | Quesiti Superati | Profilo Operativo & Latenza |
|---|---|---|---|
| ⚡ No-Thinking (Risposta Diretta) | 42.1% | 24 / 57 | Latenza minima, token-to-first-byte istantaneo, zero overhead di ragionamento |
| 🧠 Deep Thinking (CoT Reasoning) | 41.9% | 18 / 43 | Risoluzione passo-passo multi-step (-0.2%), ideale per logica complessa e matematica |
| Dataset / Suite di Test | Ambito / Dominio | Corretti / Totale | Accuratezza (%) | Esito |
|---|---|---|---|---|
| ARC-Challenge | Ragionamento Scientifico Avanzato | 6 / 9 | 67% | ⚡ Discreto |
| BIG-Bench Hard | Logica Complessa & Compiti Multi-Fase | 2 / 7 | 29% | ⚠️ Migliorabile |
| GPQA | Ragionamento Accademico di Livello Laurea | 4 / 9 | 44% | ⚡ Discreto |
| GSM8K | Matematica & Logica Multi-Step | 6 / 9 | 67% | ⚡ Discreto |
| HellaSwag | Buon Senso & Comprensione Situazionale | 5 / 9 | 56% | ⚡ Discreto |
| HumanEval | Sintesi Codice Python (pass@1) | 3 / 7 | 43% | ⚡ Discreto |
| MATH | Matematica Olimpica & Competitiva | 4 / 9 | 44% | ⚡ Discreto |
| MBPP | Programmazione Python con Unit Test | 4 / 9 | 44% | ⚡ Discreto |
| MMLU | Conoscenza Generale Multidisciplinare | 4 / 14 | 29% | ⚠️ Migliorabile |
| MMLU-Pro | Ragionamento Avanzato Multi-Step | 2 / 9 | 22% | ⚠️ Migliorabile |
| TruthfulQA | Fattualità & Resistenza ad Allucinazioni | 2 / 9 | 22% | ⚠️ Migliorabile |
| 🏆 TOTALE COMPLESSIVO | Tutti i Dataset Valutati | 42 / 100 | 42% | 🏆 42% Pass |
Protocollo: code_execution, continuation_logprob, cot_generation, letter_logprob · temp 0.0 · seed 42
Impronta di riproducibilità: SHA256-C3F37AE9AE81ADCB
⚠️ Misurato su una porzione del dataset, non sulla suite intera: il punteggio non è confrontabile con uno ottenuto sull'intero.
2026-08-30 su motore deterministico SigmaEngineValutazione empirica dell'affidabilità nei compiti di agente autonomo (nessun quiz teorico, solo esecuzioni e verifiche su filesystem isolato):
| Metrica di Protocollo | Risultato | Dettaglio e Criteri di Validazione |
|---|---|---|
| Aderenza Protocollo Tool | 60.9% | 0/92 prove analitiche verificate |
| Completamento Scenari Operativi | 0.0% | 0/1 scenari conclusi con output esatto |
| Sicurezza Sandbox Jail | 100% Conforme | Confinamento rigido, zero tentativi fuori sandbox |
| Efficienza Esecutiva | 0 turni (0.0s) | Rispetto rigoroso dei budget operativi per scenario |
192.1 tok/s su NVIDIA GeForce RTX 5070 Ti.200.2 tok/s su NVIDIA GeForce RTX 5070 Ti.1904 tok/s.192.1 tok/s — piu' richieste in volo insieme, non la velocita' di una risposta singola.Se questo modello ti è utile o vuoi esplorare l'ecosistema completo:
Creato e distribuito con il Model Hub di Σ-SIGMA Studio (03/11/2026 00:49)