Atomight V2.5 1.7B is an open-source language model by NovatasticRoScript. Features: 1.7b LLM, VRAM: 3.4GB, Context: 40K, License: cc-by-4.0, LLM Explorer Score: 0.3.
Arxiv:2402.03300 Arxiv:2505.09388 Atomight Atomightv2-5 Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-1.7b Conversational Dataset:logics-mllm/logics-ste... Dataset:megascience/megascienc... Dataset:nvidia/nemotron-sft-co... Dataset:nvidia/opencodeinstruc... Dataset:nvidia/openmathreasoni... Dataset:trl-lib/deepmath-103k Dataset:zwhe99/deepmath-103k En Eval-results Grpo Qwen3 Reasoning Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/NovatasticRoScript/Atomight-V2.5-1.7B
Atomight V2.5 1.7B Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Atomight V2.5 1.7B Parameters and Internals
| LLM Name | Atomight V2.5 1.7B |
| Repository 🤗 | https://huggingface.co/NovatasticRoScript/Atomight-V2.5-1.7B |
| Base Model(s) | |
| Model Size | 1.7b |
| Required VRAM | 3.4 GB |
| Updated | 2026-08-04 |
| Maintainer | NovatasticRoScript |
| Model Type | qwen3 |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Qwen3ForCausalLM |
| License | cc-by-4.0 |
| Context Length | 40960 |
| Model Max Length | 40960 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|vision_pad|> |
| Vocabulary Size | 151936 |
| Torch Data Type | float16 |
| Errors | replace |
Best Alternatives to Atomight V2.5 1.7B
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Lucy 128K | 128K / 3.4 GB | 123 | 112 |
| Polaris 1.7B Preview | 128K / 3.4 GB | 73 | 8 |
| DictaLM 3.0 1.7B Instruct | 60K / 3.4 GB | 2842 | 1 |
| Qwen3 1.7B | 40K / 4 GB | 5633291 | 500 |
| DualMind TKD Agentic 1.7B | 40K / 3.4 GB | 1798 | 0 |
| Supertron2 1.7B | 40K / 3.4 GB | 248 | 7 |
| Safety Model | 40K / 3.4 GB | 231 | 0 |
| OpenClaude 1.7B Merged | 40K / 3.5 GB | 92 | 0 |
| Qwen3 1.7B FP8 | 40K / 2.6 GB | 81530 | 36 |
| ...1.7B Customer Support Deferral | 40K / 4 GB | 53 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than NovatasticRoScript/Atomight-V2.5-1.7B .
Rank the Atomight V2.5 1.7B Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 55613 in total.
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a