Megachat is an open-source language model by w95. Features: 1.1b LLM, License: apache-2.0, HF Score: 34.8, LLM Explorer Score: 0.14, Arc: 30.8, HellaSwag: 54.4, MMLU: 25.6, TruthfulQA: 39.9, WinoGrande: 57, GSM8K: 1.
Base model:finetune:tinyllama/... Base model:tinyllama/tinyllama... Endpoints compatible Generated from trainer Llama Lora Pytorch Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/w95/megachat
Megachat Benchmarks
ARC: 30.8 vs 96.7 (so35)-68.1%
HellaSwag: 54.35 vs 95.3 (gpt4)-43%
MMLU: 25.55 vs 88.3 (so35)-71.1%
TruthfulQA: 39.85 vs 59 (gpt4)-32.5%
WinoGrande: 56.99 vs 87.5 (gpt4)-34.9%
GSM8K: 0.99 vs 96.4 (so35)-99%
LLME Score: 0.14323
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)
Megachat Parameters and Internals
| LLM Name | Megachat |
| Repository 🤗 | https://huggingface.co/w95/megachat |
| Base Model(s) | |
| Model Size | 1.1b |
| Required VRAM | 0 GB |
| Updated | 2026-03-29 |
| Maintainer | w95 |
| Model Files | |
| Model Architecture | AutoModelForCausalLM |
| License | apache-2.0 |
| Is Biased | none |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | </s> |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | q_proj|v_proj |
| LoRA Alpha | 16 |
| LoRA Dropout | 0.05 |
| R Param | 8 |
Best Alternatives to Megachat
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Tiny Llama Miniguanaco | 2K / 2.2 GB | 41 | 1 |
| MindEase Assistant | 0K / 2.2 GB | 8 | 0 |
| Tiny Darkllama DPO | 0K / 0.6 GB | 7 | 2 |
| Cookgptlama | 0K / 0 GB | 6092 | 3 |
| Tiny Llama Sql DataAdmin | 0K / 0 GB | 5 | 0 |
| Tiny Llama Agri Bot | 0K / 0 GB | 5 | 0 |
| ...ny Lllama SqlQuries Fine Tuned | 0K / 0 GB | 5 | 0 |
| New Tiny Llama1B 12K | 0K / 2.2 GB | 8 | 0 |
| Manovyadh 1.1B V1 | 0K / 0.1 GB | 1 | 1 |
| Tinyllava V1.0 1.1B Lora | 0K / 0.2 GB | 8 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than w95/megachat .
Rank the Megachat 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? 52721 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a