LLM EXPLORER 59,516 MODELS INDEXED

Dolphin 2.9.2 Phi 3 Medium by cognitivecomputations

By cognitivecomputations · 9292 downloads
HF Repo ↗ Copy URL

Dolphin 2.9.2 Phi 3 Medium is an open-source language model by cognitivecomputations. Features: LLM, VRAM: 28GB, Context: 4K, License: mit, Instruction-Based, LLM Explorer Score: 0.24.

  Autotrain compatible Base model:finetune:unsloth/ph... Base model:unsloth/phi-3-mini-...   Conversational Dataset:cognitivecomputations/... Dataset:cognitivecomputations/... Dataset:cognitivecomputations/... Dataset:cognitivecomputations/...   Dataset:internlm/agent-flan Dataset:m-a-p/codefeedback-fil... Dataset:microsoft/orca-math-wo...   Dataset:teknium/openhermes-2.5   En   Endpoints compatible   Instruct   Mistral   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/cognitivecomputations/dolphin-2.9.2-Phi-3-Medium 

Dolphin 2.9.2 Phi 3 Medium Parameters and Internals

Model Type 
instruction, conversational, coding
Additional Notes 
The model has been filtered to remove alignment and bias.
Supported Languages 
en (English), proficiency (Unknown)
Training Details 
Data Sources:
cognitivecomputations/Dolphin-2.9.2, teknium/OpenHermes-2.5, m-a-p/CodeFeedback-Filtered-Instruction, cognitivecomputations/dolphin-coder, cognitivecomputations/samantha-data, microsoft/orca-math-word-problems-200k, internlm/Agent-FLAN, cognitivecomputations/SystemChat-2.0
Methodology:
qLoRA fine-tuning with 4k sequence length
Context Length:
4000
Training Time:
3.5 days
Hardware Used:
8xL40S node provided by Crusoe Cloud
Model Architecture:
Based on Phi-3-Medium
Input Output 
Input Format:
ChatML prompt template
Accepted Modalities:
text
Output Format:
text outputs
Performance Tips:
Implement your own alignment layer before exposing the model as a service.
LLM NameDolphin 2.9.2 Phi 3 Medium
Repository 🤗https://huggingface.co/cognitivecomputations/dolphin-2.9.2-Phi-3-Medium 
Base Model(s)  Phi 3 Mini 4K Instruct   unsloth/Phi-3-mini-4k-instruct
Required VRAM28 GB
Updated2025-07-13
Maintainercognitivecomputations
Model Typemistral
Instruction-BasedYes
Model Files  4.9 GB: 1-of-6   5.0 GB: 2-of-6   4.9 GB: 3-of-6   5.0 GB: 4-of-6   5.0 GB: 5-of-6   3.2 GB: 6-of-6
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licensemit
Context Length4096
Model Max Length4096
Transformers Version4.40.2
Tokenizer ClassLlamaTokenizer
Padding Token<|placeholder6|>
Vocabulary Size32064
Torch Data Typebfloat16

Quantized Models of the Dolphin 2.9.2 Phi 3 Medium

Model
Likes
Downloads
VRAM
... Phi 3 Medium HQQ 4bit Smashed057 GB
... Phi 3 Medium HQQ 2bit Smashed054 GB
... Phi 3 Medium HQQ 1bit Smashed052 GB
... Phi 3 Medium AWQ 4bit Smashed097 GB
... Phi 3 Medium Bnb 4bit Smashed058 GB

Best Alternatives to Dolphin 2.9.2 Phi 3 Medium

Best Alternatives
Context / RAM
Downloads
Likes
Krutrim 2 Instruct1000K / 49.3 GB86638
Mistral Instruct SCQA32K / 5 GB80
Dolphin 2.9.2 Phi 3 Medium4K / 28 GB820926
Lora Phi3 Mini4K / 5 GB70
...eZephir Sft Instruct Ead 16bit32K / 14.4 GB560
... Instruct Unsloth Merged 16bit4K / 7.6 GB61
... Mini 4K Instruct Bnb 4bit Ita4K / 7.6 GB11810
... Mini 4K Instruct Bnb 4bit Ita4K / 7.6 GB15270
...ini New Model With Lora Merged4K / 7.6 GB160
Phired4K / 7.6 GB120