LLM EXPLORER 59,420 MODELS INDEXED

14B by CausalLM

By CausalLM · 222 downloads
HF Repo ↗ Copy URL

14B is an open-source language model by CausalLM. Features: 14b LLM, VRAM: 28.4GB, Context: 8K, License: wtfpl, Instruction-Based, LLM Explorer Score: 0.17, Arc: 56.7, HellaSwag: 79.1, MMLU: 65.9, GSM8K: 58.6.

  Causallm   Dataset:baai/coig   Dataset:fnlp/moss-003-sft-data Dataset:garage-baind/open-plat... Dataset:jondurbin/airoboros-3.... Dataset:josephuscheung/guanaco...   Dataset:ldjnr/puffin Dataset:liuhaotian/llava-instr...   Dataset:liwu/mnbvc   Dataset:meta-math/metamathqa Dataset:milashkaarshif/moegirl...   Dataset:open-orca/openorca   Dataset:openbmb/llava zh   Dataset:ryokoai/fandom23k   Dataset:ryokoai/sharegpt52k   Dataset:stingning/ultrachat   Dataset:teknium/openhermes Dataset:tigerresearch/tigerbot...   Dataset:wiki lingua   Dataset:wikipedia Dataset:wizardlm/wizardlm evol...   En   Endpoints compatible   Instruct   Llama   Llama2   Pytorch   Qwen   Region:us   Sharded   Zh
Model Card on HF πŸ€—: https://huggingface.co/CausalLM/14B 

14B Parameters and Internals

Model Type 
text-generation, causallm
Use Cases 
Areas:
Research, Commercial applications
Primary Use Cases:
Speculative sampling
Limitations:
May produce hallucinations or unreliable outputs, Contains objectionable content, pornography, violence, and offensive language
Considerations:
Conduct own safety checks, filter keywords
Additional Notes 
The model is not recommended for quantization, but rather using smaller size models like 7B.
Supported Languages 
en (English), zh (Chinese)
Training Details 
Data Sources:
JosephusCheung/GuanacoDataset, Open-Orca/OpenOrca, stingning/ultrachat, meta-math/MetaMathQA, liuhaotian/LLaVA-Instruct-150K, jondurbin/airoboros-3.1, WizardLM/WizardLM_evol_instruct_V2_196k, RyokoAI/ShareGPT52K, RyokoAI/Fandom23K, milashkaarshif/MoeGirlPedia_wikitext_raw_archive, wikipedia, wiki_lingua, fnlp/moss-003-sft-data, garage-bAInd/Open-Platypus, LDJnr/Puffin, openbmb/llava_zh, BAAI/COIG, TigerResearch/tigerbot-zhihu-zh-10k, liwu/MNBVC, teknium/openhermes
Data Volume:
1.3B tokens
Methodology:
Manual or synthetic rewrites, augmented text training, synthetic Wikipedia conversation dataset
Model Architecture:
Identical to LLaMA2, using same attention calculation method
Input Output 
Input Format:
[chatml](https://github.com/openai/openai-python/blob/main/chatml.md)
Accepted Modalities:
text
Output Format:
text
Performance Tips:
If VRAM is insufficient, use the 7B model instead of the quantized version.
Release Notes 
Version:
14B-DPO-alpha
Date:
Dec 3, 2023
Notes:
Outperforms Zephyr-Ξ² in MT-Bench.
LLM Name14B
Repository πŸ€—https://huggingface.co/CausalLM/14B 
Model Size14b
Required VRAM28.4 GB
Updated2026-08-25
MaintainerCausalLM
Model Typellama
Instruction-BasedYes
Model Files  10.0 GB: 1-of-3   10.0 GB: 2-of-3   8.4 GB: 3-of-3
Supported Languagesen zh
Model ArchitectureLlamaForCausalLM
Licensewtfpl
Context Length8192
Model Max Length8192
Transformers Version4.35.0.dev0
Tokenizer ClassGPT2Tokenizer
Vocabulary Size152064
Torch Data Typebfloat16

Quantized Models of the 14B

Model
Likes
Downloads
VRAM
CausalLM 14B GGUF19642628 GB
CausalLM 14B EXL2388 GB
CausalLM 14B GPTQ20599 GB
CausalLM 14B AWQ10489 GB

Best Alternatives to 14B

Best Alternatives
Context / RAM
Downloads
Likes
...struct 2512 Llamafied TextOnly256K / 27 GB93
...hocolatine 2 14B Instruct V2.016K / 29.4 GB116
...colatine 2 14B Instruct V2.0.116K / 29.4 GB301
CausalLM 14B EXL28K / 8.5 GB83
CausalLM 14B GPTQ8K / 9.7 GB5920
CausalLM 14B AWQ8K / 9.7 GB4810
Note: green Score (e.g. "73.2") means that the model is better than CausalLM/14B.