LLM EXPLORER 61,491 MODELS INDEXED

Bagel 34B V0.2 by jondurbin

By jondurbin · 259 downloads
HF Repo ↗ Copy URL

Bagel 34B V0.2 is an open-source language model by jondurbin. Features: 34b LLM, VRAM: 68.7GB, Context: 195K, License: apache-2.0, LLM Explorer Score: 0.18, Arc: 68.8, HellaSwag: 83.7, MMLU: 76.5, GSM8K: 46.2.

  Conversational   Dataset:ai2 arc Dataset:allenai/ultrafeedback ...   Dataset:boolq   Dataset:cais/mmlu   Dataset:cakiki/rosetta-code   Dataset:codeparrot/apps   Dataset:datasets/winogrande   Dataset:drop   Dataset:facebook/belebele   Dataset:intel/orca dpo pairs Dataset:jondurbin/cinematika-v... Dataset:jondurbin/truthy-dpo-v...   Dataset:julielab/emobank   Dataset:kingbri/pippa-sharegpt   Dataset:ldjnr/capybara   Dataset:lmsys/lmsys-chat-1m Dataset:migtissera/synthia-v1.... Dataset:muennighoff/natural-in...   Dataset:nvidia/helpsteer   Dataset:open-orca/slimorca   Dataset:openbookqa   Dataset:piqa   Dataset:spider   Dataset:squad v2 Dataset:squish42/bluemoon-fand...   Dataset:tiger-lab/mathinstruct   Dataset:unalignment/spicy-3.1 Dataset:unalignment/toxic-dpo-... Dataset:vezora/tested-22k-pyth...   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/jondurbin/bagel-34b-v0.2 

Bagel 34B V0.2 Parameters and Internals

Model Type 
Text Generation
Training Details 
Data Sources:
ai2_arc, unalignment/spicy-3.1, codeparrot/apps, facebook/belebele, boolq, jondurbin/cinematika-v0.1, drop, lmsys/lmsys-chat-1m, TIGER-Lab/MathInstruct, cais/mmlu, Muennighoff/natural-instructions, openbookqa, piqa, Vezora/Tested-22k-Python-Alpaca, cakiki/rosetta-code, Open-Orca/SlimOrca, spider, squad_v2, migtissera/Synthia-v1.3, datasets/winogrande, nvidia/HelpSteer, Intel/orca_dpo_pairs, unalignment/toxic-dpo-v0.1, jondurbin/truthy-dpo-v0.1, allenai/ultrafeedback_binarized_cleaned, Squish42/bluemoon-fandom-1-1-rp-cleaned, LDJnr/Capybara, JULIELab/EmoBank, kingbri/PIPPA-shareGPT
Methodology:
SFT phase using a variety of prompt formats: vicuna, llama-2, alpaca, and chat-ml (sorta).
Model Architecture:
unknown
Input Output 
Input Format:
vicuna, llama-2, alpaca, chat-ml
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Recommended to run for 1 epoch or 0.75 epoch, using a relatively low learning rate.
LLM NameBagel 34B V0.2
Repository 🤗https://huggingface.co/jondurbin/bagel-34b-v0.2 
Model Size34b
Required VRAM68.7 GB
Updated2026-08-06
Maintainerjondurbin
Model Typellama
Model Files  4.0 GB: 1-of-18   3.9 GB: 2-of-18   3.9 GB: 3-of-18   3.9 GB: 4-of-18   3.9 GB: 5-of-18   3.9 GB: 6-of-18   3.9 GB: 7-of-18   3.9 GB: 8-of-18   3.9 GB: 9-of-18   3.9 GB: 10-of-18   3.9 GB: 11-of-18   3.9 GB: 12-of-18   3.9 GB: 13-of-18   3.9 GB: 14-of-18   3.9 GB: 15-of-18   3.9 GB: 16-of-18   3.9 GB: 17-of-18   2.3 GB: 18-of-18
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length200000
Model Max Length200000
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size64000
Torch Data Typebfloat16

Quantized Models of the Bagel 34B V0.2

Model
Likes
Downloads
VRAM
Luminex 34B V0.1 EXL2 8.0bpw1134 GB
Smaug 34B V0.1 2.65bpw H6 EXL21412 GB
Smaug 34B V0.1 2.4bpw H6 EXL21211 GB
Smaug 34B V0.1 4.0bpw H6 EXL23318 GB
Smaug 34B V0.1 8.0bpw H8 EXL21834 GB
Smaug 34B V0.1 3.0bpw H6 EXL22413 GB
Smaug 34B V0.1 AWQ2319 GB
Smaug 34B V0.1 6.0bpw H6 EXL21626 GB
Smaug 34B V0.1 4.65bpw H6 EXL21520 GB
Smaug 34B V0.1 5.0bpw H6 EXL21322 GB

Best Alternatives to Bagel 34B V0.2

Best Alternatives
Context / RAM
Downloads
Likes
Bagel Hermes 34B Slerp195K / 68.9 GB83751
34B Beta195K / 69.2 GB867067
Smaug 34B V0.1195K / 69.2 GB808264
Casual Magnum 34B195K / 68.8 GB41
Yi 34B 200K195K / 68.9 GB9133320
Yi 34B 200K AEZAKMI V2195K / 69.2 GB5712
Faro Yi 34B195K / 69.2 GB82086
Bagel DPO 34B V0.5195K / 68.7 GB804017
Smaug 34B V0.1 ExPO195K / 69.2 GB78120
Luminex 34B V0.1195K / 68.9 GB85668
Note: green Score (e.g. "73.2") means that the model is better than jondurbin/bagel-34b-v0.2.