LLM EXPLORER 59,420 MODELS INDEXED

Deep Ft5 Grp 16bit by CoexistAI

By CoexistAI · 27 downloads
HF Repo ↗ Copy URL

Deep Ft5 Grp 16bit is an open-source language model by CoexistAI. Features: 1.5b LLM, VRAM: 3.5GB, Context: 128K, License: apache-2.0, Quantized, LLM Explorer Score: 0.17.

  4bit   Conversational   En   Endpoints compatible   Grpo   Pytorch   Quantized   Qwen2   Region:us   Trl   Unsloth
Model Card on HF 🤗: https://huggingface.co/CoexistAI/deep_ft5_grp_16bit 

Deep Ft5 Grp 16bit Parameters and Internals

LLM NameDeep Ft5 Grp 16bit
Repository 🤗https://huggingface.co/CoexistAI/deep_ft5_grp_16bit 
Base Model(s)  unsloth/deepseek-r1-distill-qwen-1.5b-bnb-4bit   unsloth/deepseek-r1-distill-qwen-1.5b-bnb-4bit
Model Size1.5b
Required VRAM3.5 GB
Updated2026-08-25
MaintainerCoexistAI
Model Typeqwen2
Model Files  3.5 GB
Supported Languagesen
Quantization Type4bit
Model ArchitectureQwen2ForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.48.3
Tokenizer ClassLlamaTokenizer
Padding Token<|vision_pad|>
Vocabulary Size151936
Torch Data Typefloat16

Quantized Models of the Deep Ft5 Grp 16bit

Model
Likes
Downloads
VRAM
Deep Ft6 Grp 16bit0143 GB

Best Alternatives to Deep Ft5 Grp 16bit

Best Alternatives
Context / RAM
Downloads
Likes
VibeThinker 1.5B Mlx 4bit128K / 0.9 GB105526
...ill Qwen 1.5B Unsloth Bnb 4bit128K / 1.8 GB1106616
Eval Model128K / 7.1 GB91
DeepSeek R1 ForumNXT128K / 3.5 GB80
Research PathfinderAI128K / 3.5 GB190
...en 1.5B Trained Alvin0616 Fork128K / 3.5 GB140
... R1 Distill Qwen 1.5B Bnb 4bit128K / 1.6 GB200811
...pSeek R1 Distill Qwen 1.5B 500128K / 1.7 GB60
Umbr0x 1.5B V3.1 16bit 2128K / 15.2 GB190
Qwen2 1.5B SFT 16bit Alpha128K / 3.1 GB80
Note: green Score (e.g. "73.2") means that the model is better than CoexistAI/deep_ft5_grp_16bit.