Instructions to use NANI-Nithin/manaca-1b-base-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use NANI-Nithin/manaca-1b-base-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

Use Docker

docker model run hf.co/NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use NANI-Nithin/manaca-1b-base-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NANI-Nithin/manaca-1b-base-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NANI-Nithin/manaca-1b-base-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

Ollama
How to use NANI-Nithin/manaca-1b-base-GGUF with Ollama:
```
ollama run hf.co/NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use NANI-Nithin/manaca-1b-base-GGUF with Docker Model Runner:
```
docker model run hf.co/NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M
```

Lemonade

How to use NANI-Nithin/manaca-1b-base-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.manaca-1b-base-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

manaca-1b-base GGUF

GGUF quantizations of menezesbruno/manaca-1b-base, covering 30 files (27.0 GB total).

Files

File	Quant	Size	Notes
manaca-1b-base-BF16.gguf	`BF16`	3.21 GB	Full precision source. Every quant below is cut from this file.
manaca-1b-base-Q8_0.gguf	`Q8_0`	1.71 GB	Effectively lossless. Use when disk and RAM are not the constraint.
manaca-1b-base-Q6_K.gguf	`Q6_K`	1.32 GB	Near-lossless; the last stop before quality becomes measurable.
manaca-1b-base-Q5_K_M.gguf	`Q5_K_M`	1.15 GB	Very good quality, noticeably smaller than Q6_K.
manaca-1b-base-Q5_K_S.gguf	`Q5_K_S`	1.12 GB	Slightly smaller than Q5_K_M for a slight quality cost.
manaca-1b-base-Q5_1.gguf	`Q5_1`	1.22 GB	Legacy. Prefer Q5_K_M.
manaca-1b-base-Q5_0.gguf	`Q5_0`	1.12 GB	Legacy. Prefer Q5_K_M.
manaca-1b-base-Q4_K_M.gguf	`Q4_K_M`	0.99 GB	The usual default. Best quality-per-byte for most people.
manaca-1b-base-Q4_K_S.gguf	`Q4_K_S`	0.94 GB	A little smaller than Q4_K_M, a little worse.
manaca-1b-base-IQ4_NL.gguf	`IQ4_NL`	0.94 GB	Non-linear 4-bit; good on hardware without fast K-quant kernels.
manaca-1b-base-IQ4_XS.gguf	`IQ4_XS`	0.90 GB	Best sub-4.5bpw option; usually beats Q4_K_S at a smaller size.
manaca-1b-base-Q4_1.gguf	`Q4_1`	1.03 GB	Legacy. Prefer Q4_K_M.
manaca-1b-base-Q4_0.gguf	`Q4_0`	0.94 GB	Legacy round-to-nearest. Prefer Q4_K_M unless a runtime needs this.
manaca-1b-base-Q3_K_L.gguf	`Q3_K_L`	0.87 GB	Small, with real quality loss. Usable when RAM is tight.
manaca-1b-base-Q3_K_M.gguf	`Q3_K_M`	0.81 GB	Smaller again; noticeable degradation.
manaca-1b-base-IQ3_M.gguf	`IQ3_M`	0.76 GB	Strong at ~3.7bpw, clearly better than Q3_K_M.
manaca-1b-base-IQ3_S.gguf	`IQ3_S`	0.74 GB	Slightly smaller than IQ3_M.
manaca-1b-base-Q3_K_S.gguf	`Q3_K_S`	0.74 GB	Aggressive. Prefer IQ3_M at a similar size.
manaca-1b-base-IQ3_XS.gguf	`IQ3_XS`	0.71 GB	Aggressive but coherent.
manaca-1b-base-IQ3_XXS.gguf	`IQ3_XXS`	0.66 GB	Very aggressive; imatrix carries it.
manaca-1b-base-Q2_K.gguf	`Q2_K`	0.64 GB	Very small, heavily degraded. For experimentation.
manaca-1b-base-IQ2_M.gguf	`IQ2_M`	0.60 GB	The smallest size most people find usable.
manaca-1b-base-Q2_K_S.gguf	`Q2_K_S`	0.61 GB	Smaller than Q2_K, requires the imatrix.
manaca-1b-base-IQ2_S.gguf	`IQ2_S`	0.56 GB	Below the usual usability line.
manaca-1b-base-IQ2_XS.gguf	`IQ2_XS`	0.53 GB	Experimental.
manaca-1b-base-IQ2_XXS.gguf	`IQ2_XXS`	0.49 GB	Experimental.
manaca-1b-base-Q2_0.gguf	`Q2_0`	0.56 GB	Extreme, group-64. Included for completeness.
manaca-1b-base-IQ1_M.gguf	`IQ1_M`	0.44 GB	Extreme. Expect substantial degradation.
manaca-1b-base-IQ1_S.gguf	`IQ1_S`	0.41 GB	Extreme. Expect substantial degradation.
manaca-1b-base-Q1_0.gguf	`Q1_0`	0.31 GB	Extreme. Included for completeness.

Which one should I download?

Pick the largest file that leaves a couple of gigabytes of headroom on the device you will run it on — the model has to fit in RAM (or VRAM, if you are offloading) alongside the KV cache and the OS.

Plenty of memory: Q6_K or Q8_0.
The usual choice: Q4_K_M.
Tight on memory: IQ4_XS, then IQ3_M, then IQ2_M.
The IQ* files are imatrix-guided and generally beat a Q* file of similar size, at the cost of slightly slower inference on some hardware.

Quantization details

Importance matrix computed with llama-imatrix over 500 rows of Salesforce/wikitext (wikitext-2-raw-v1).
The matrix was computed on the BF16 weights.
K-quants below 6 bit and the whole IQ set are imatrix-guided. Q4_0/Q4_1/Q5_0/Q5_1 are legacy round-to-nearest and ignore it; Q6_K/Q8_0 are near-lossless and do not need it.
All files are cut from the same BF16 GGUF, so differences between them are quantization only.

Usage

llama-cli -hf NANI-Nithin/manaca-1b-base-GGUF:Q4_K_M -p "Hello"

Or download one file and point at it directly:

huggingface-cli download NANI-Nithin/manaca-1b-base-GGUF manaca-1b-base-Q4_K_M.gguf --local-dir .
llama-cli -m manaca-1b-base-Q4_K_M.gguf -p "Hello"

Quantized with llama.cpp by AgentQuantix on 2026-09-05. Licensing follows the base model.

Downloads last month: 837

GGUF

Model size

2B params

Architecture

llama

Hardware compatibility

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for NANI-Nithin/manaca-1b-base-GGUF

Base model

menezesbruno/manaca-1b-base

Quantized

(5)

this model