Instructions to use robolamp/AliceAI-Foundation-80B-A3B-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 robolamp/AliceAI-Foundation-80B-A3B-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 robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf robolamp/AliceAI-Foundation-80B-A3B-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 robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M

Use Docker

docker model run hf.co/robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "robolamp/AliceAI-Foundation-80B-A3B-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": "robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M

Ollama
How to use robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF with Ollama:
```
ollama run hf.co/robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF with Docker Model Runner:
```
docker model run hf.co/robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M
```

Lemonade

How to use robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.AliceAI-Foundation-80B-A3B-Base-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

AliceAI-Foundation-80B-A3B-Base-GGUF

GGUF port of yandex/AliceAI-Foundation-80B-A3B-Base: MoE, 80B total / 3B active parameters, tokenizer embedded in every file. This is a base model — no chat template, not instruction-tuned; it continues raw text.

Requires a llama.cpp fork. The alice_ai architecture is not in upstream llama.cpp; a stock build fails with unknown model architecture. Use robolamp/llama.cpp at commit 9a9ca8b. This notice will be removed once the port lands upstream.

UPD: now supports MTP, recommended --spec-draft-n-max for Q4 is 2, for Q8 use 3.

Variants

Variant	Size	TVD vs f32¹	Head²	Ties³	Status
`BF16`	159.4 GB	—	—	—	reference
`Q8_0` (2 shards)	84.8 GB	2.71e-02	100%	1	✅
`Q6_K`	65.5 GB	3.42e-02	100%	6	✅
`Q5_K_M`	56.7 GB	4.04e-02	100%	7	✅
`Q5_K_S`	55.0 GB	4.18e-02	100%	8	✅
`Q4_K_M`	48.4 GB	8.13e-02	100%	9	✅ recommended
`Q4_K_S`	45.5 GB	9.21e-02	100%	11	✅
`Q3_K_L`	41.2 GB	9.51e-02	100%	19	✅
`Q3_K_S`	34.5 GB	1.19e-01	50%	18	not recommended
`Q2_K`	29.1 GB	1.39e-01	33%	13	not recommended

Full file names: AliceAI-Foundation-80B-A3B-Base-<variant>.gguf.

¹ Total variation distance between next-token distributions and an f32 reference on a held-out prompt set; lower is better. Measured on the full model for Q8_0, on a truncated probe for the other rows. The port itself matches the HF f32 reference within 3.48e-05 TVD (100% head).

² Share of mass-carrying tokens (p ≥ 0.01) whose top choice is identical.

³ Number of forced chosen-token ties.

Not published: Q3_K_M. Head overlap drops to 50% and greedy generation degenerates on some prompts.
No IQ* / imatrix variants: no importance matrix exists for this model yet.

Conversion notes

The tokenizer follows transformers, not the SentencePiece proto (they disagree on leading whitespace), via a new optional key tokenizer.ggml.spm_prepend_hf (default off, other models unaffected). 1028 USER_DEFINED vocabulary pieces are promoted, so control tokens like [COT_ENABLE] are not fragmented.

License

Derivative of yandex/AliceAI-Foundation-80B-A3B-Base — the original license applies. Architecture port, conversion and verification: this repository's author, Yulia Yakovleva.

Downloads last month: 2,726

GGUF

Model size

81B params

Architecture

alice_ai

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for robolamp/AliceAI-Foundation-80B-A3B-Base-GGUF

Base model

yandex/AliceAI-Foundation-80B-A3B-Base

Quantized

(8)

this model