LLM EXPLORER 58,975 MODELS INDEXED

Nemotron Research GooseReason 4B Instruct by nvidia

By nvidia · 83 downloads
HF Repo ↗ Copy URL

Nemotron Research GooseReason 4B Instruct is an open-source language model by nvidia. Features: 4b LLM, VRAM: 17.6GB, Context: 256K, License: cc-by-nc-4.0, Instruction-Based, LLM Explorer Score: 0.22.

  Arxiv:2505.24864   Arxiv:2601.22975 Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-4b-instr...   Code   Conversational   En   Endpoints compatible   Instruct   Math   Nvidia   Qwen3   Reasoning   Region:us   Rlvr   Safetensors   Sharded   Stem   Tensorflow
Model Card on HF 🤗: https://huggingface.co/nvidia/Nemotron-Research-GooseReason-4B-Instruct 

Nemotron Research GooseReason 4B Instruct Parameters and Internals

LLM NameNemotron Research GooseReason 4B Instruct
Repository 🤗https://huggingface.co/nvidia/Nemotron-Research-GooseReason-4B-Instruct 
Base Model(s)  Qwen3 4B Instruct 2507   Qwen/Qwen3-4B-Instruct-2507
Model Size4b
Required VRAM17.6 GB
Updated2026-07-19
Maintainernvidia
Model Typeqwen3
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   4.9 GB: 2-of-4   5.0 GB: 3-of-4   2.7 GB: 4-of-4
Supported Languagesen
Model ArchitectureQwen3ForCausalLM
Licensecc-by-nc-4.0
Context Length262144
Model Max Length262144
Transformers Version4.56.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Errorsreplace

Best Alternatives to Nemotron Research GooseReason 4B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
FastContext 1.0 4B SFT256K / 8.1 GB5735357
Qwen3 4B Instruct 2507256K / 8.1 GB3109972911
Fable Traces256K / 8.1 GB375210
FastContext 1.0 4B RL256K / 8.1 GB455961
Qwen3 4B Instruct 2507 FP8256K / 5.2 GB110074478
...pt Full Qwen3 4b Instruct 2507256K / 8 GB3651
Nexa AI 4B Instruct256K / 8 GB5762
Neuron 4B Instruct256K / 8.1 GB3131
Agents K1256K / 8.8 GB109828
FastContext 1.0 4B SFT256K / 8.1 GB27318
Note: green Score (e.g. "73.2") means that the model is better than nvidia/Nemotron-Research-GooseReason-4B-Instruct.