LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Stable Code 3B by stabilityai

 »  All LLMs  »  stabilityai  »  Stable Code 3B   URL Share it on

  6677   

Stable Code 3B is an open-source language model by stabilityai. Features: 3b LLM, VRAM: 5.6GB, Context: 16K, License: other, Quantized, HF Score: 41.5, LLM Explorer Score: 0.22, Arc: 35.3, HellaSwag: 57.5, MMLU: 37.5, TruthfulQA: 41.6, WinoGrande: 57.7, GSM8K: 19.6, HumanEval: 30.7.

  Arxiv:1910.02054   Arxiv:2104.09864   Arxiv:2204.06745   Arxiv:2305.06161   Arxiv:2307.09288   Arxiv:2309.12284   Arxiv:2310.10631   Code   Dataset:bigcode/commitpackft   Dataset:bigcode/starcoderdata Dataset:bigcode/the-stack-gith... Dataset:eleutherai/proof-pile-...   Dataset:meta-math/metamathqa Dataset:tiiuae/falcon-refinedw...   En   Endpoints compatible   Gguf   Model-index   Q5   Quantized   Region:us   Safetensors   Sharded   Stablelm   Tensorflow
Model Card on HF 🤗: https://huggingface.co/stabilityai/stable-code-3b 

Stable Code 3B Benchmarks

ARC: 35.32 vs 96.7 (so35)-63.5%
HellaSwag: 57.45 vs 95.3 (gpt4)-39.7%
MMLU: 37.53 vs 88.3 (so35)-57.5%
TruthfulQA: 41.6 vs 59 (gpt4)-29.5%
WinoGrande: 57.7 vs 87.5 (gpt4)-34.1%
GSM8K: 19.56 vs 96.4 (so35)-79.7%
HumanEvalA: 19.06
Python Eva: 30.72 vs 92 (so35)-66.6%
Java Eval: 28.75
Javascript: 31.64
CPP Eval: 29.42
PHP Eval: 23.68
Julia Eval: 21.41
D Eval: 10.09
Lua Eval: 17.54
R Eval: 13.37
Rust Eval: 22.15
LLME Score: 0.22249
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Stable Code 3B (stabilityai/stable-code-3b)
🌟 Advertise your project 🚀

Stable Code 3B Parameters and Internals

Model Type 
causal-lm, code, text-generation
Use Cases 
Areas:
Research, Commercial applications
Limitations:
The model may exhibit unreliable or undesirable behavior without fine-tuning., Content may be offensive or inappropriate after pre-training.
Considerations:
Evaluation and fine-tuning required prior to deployment for safe performance.
Supported Languages 
English (fluent), Code (specialized in 18 programming languages)
Training Details 
Data Sources:
tiiuae/falcon-refinedweb, bigcode/the-stack-github-issues, bigcode/commitpackft, bigcode/starcoderdata, EleutherAI/proof-pile-2, meta-math/MetaMathQA
Data Volume:
1.3 trillion tokens
Methodology:
Pre-trained using a mixture of large-scale datasets, applying rotary position embeddings for improved throughput.
Context Length:
16384
Hardware Used:
256 NVIDIA A100 40GB GPUs
Model Architecture:
Decoder-only transformer with modifications similar to LLaMA architecture, utilizing Rotary Position Embeddings.
LLM NameStable Code 3B
Repository 🤗https://huggingface.co/stabilityai/stable-code-3b 
Model Size3b
Required VRAM5.6 GB
Updated2026-04-04
Maintainerstabilityai
Model Typestablelm
Model Files  5.0 GB: 1-of-2   0.6 GB: 2-of-2   2.0 GB   2.3 GB   5.6 GB
Supported Languagesen
GGUF QuantizationYes
Quantization Typegguf|q5|q5_k
Model ArchitectureStableLmForCausalLM
Licenseother
Context Length16384
Model Max Length16384
Transformers Version4.38.0
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50304
Torch Data Typebfloat16

Quantized Models of the Stable Code 3B

Model
Likes
Downloads
VRAM
Stable Code 3B GGUF309711 GB
Stable Code 3B GPTQ12131 GB

Best Alternatives to Stable Code 3B

Best Alternatives
Context / RAM
Downloads
Likes
Stable Cypher Instruct 3B16K / 1.7 GB115828
Stable Code Instruct 3B16K / 5.6 GB1410182
Stable Cypher Instruct 3B16K / 1.7 GB26326
...struct 3B Spider 1500 Steps Q416K / 1.7 GB150
NSFW 3B4K / 10.6 GB15664194
HELVETE 3B4K / 10.6 GB526199
HelpingAI 3B Chat4K / 10.6 GB843
Rohit Sharma4K / 10.6 GB491
Stable Code Instruct 3B 4bit16K / 1.8 GB525
Stable Code 3B 4bit16K / 1.8 GB111

Rank the Stable Code 3B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 52758 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer