LLM EXPLORER 58,178 MODELS INDEXED

Mankei 1B Chat by keyvan-ai

By keyvan-ai · 0 downloads
HF Repo ↗ Copy URL

Mankei 1B Chat is an open-source language model by keyvan-ai. Features: 1b LLM, VRAM: 0.5GB, Context: 4K, License: apache-2.0, Quantized, LLM Explorer Score: 0.26.

  Conversational   Cpu Dataset:coral-nlp/german-commo...   Dataset:epfml/fineweb2-hq   Dataset:huggingfacefw/finepdfs Dataset:huggingfacefw/fineweb-...   Dataset:huggingfacetb/finemath Dataset:open-web-math/open-web...   Dataset:wikimedia/wikipedia   De   Deutsch   En   Endpoints compatible   Experimental   From-scratch   German   Gguf   Llama   Q3   Quantized   Region:us   Safetensors   Small-language-model
Model Card on HF 🤗: https://huggingface.co/keyvan-ai/Mankei-1B-Chat 

Mankei 1B Chat Parameters and Internals

LLM NameMankei 1B Chat
Repository 🤗https://huggingface.co/keyvan-ai/Mankei-1B-Chat 
Model Size1b
Required VRAM0.5 GB
Updated2026-08-12
Maintainerkeyvan-ai
Model Typellama
Model Files  2.0 GB   0.5 GB   0.7 GB   0.8 GB   0.9 GB   1.1 GB   2.0 GB
Supported Languagesde en
GGUF QuantizationYes
Quantization Typeq3|gguf|q4_k|q5_k
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version5.8.1
Padding Token<|end_of_text|>
Vocabulary Size32768

Best Alternatives to Mankei 1B Chat

Best Alternatives
Context / RAM
Downloads
Likes
...pstream Minicpm5 1B Forecaster128K / 0.7 GB2960
Llama 3.2 1B Aimv2 V2128K / 0.8 GB620
Llama 3.2 1B Sarcasm Rewriter128K / 2.5 GB120
Llama 3.2 1B Instruct GGUF128K / 0.4 GB1914469
Llama 3.2 SUN 1B Chat128K / 1.6 GB342
Llama 3.2 1B128K / 0.9 GB1020
Regex AI Llama 3.2 1B128K / 0.8 GB1151
Llama 3 Fine Tuned C128K / 1.3 GB140
Rodin 1B2K / 0.9 GB3112
Test Gguf Trainer2K / 0 GB171
Note: green Score (e.g. "73.2") means that the model is better than keyvan-ai/Mankei-1B-Chat.