Instructions to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF")
model = AutoModelForCausalLM.from_pretrained("AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-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 AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-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 AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-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 AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-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 AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M

Use Docker

docker model run hf.co/AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M

SGLang

How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with Ollama:
```
ollama run hf.co/AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with Docker Model Runner:
```
docker model run hf.co/AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M
```

Lemonade

How to use AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

Model Card for Tiny Aya L2-Thinker

Model Summary

Tiny-Aya-3.3B-L2-Thinker-GGUF CohereLabs/tiny-aya-l2-thinker

Cohere Labs Tiny Aya L2-Thinker is an open-weights research release of a 3.35 billion parameter multilingual reasoning model optimized to think in the same language as the user prompt before writing the final answer. It is trained to support in-language reasoning for 44 languages plus English, with coverage extending to 20+ more through additional non-reasoning instruction data. The model is designed to support mathematics, science, and general reasoning tasks, as well as instruction following and multilingual open-ended generation.

This is a different model from Tiny Aya En-Thinker, which thinks in English.

Developed by: Cohere and Cohere Labs

Point of Contact: Cohere Labs
License: CC-BY-NC, requires also adhering to Cohere Lab's Acceptable Use Policy
Model: Tiny-Aya-3.3B-L2-Thinker-GGUF
Model Size: 3.35B
Context Length: 32K (input + output)

For the broader Tiny Aya family, see tiny-aya-global, tiny-aya-base, and the Tiny Aya collection.

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "CohereLabs/tiny-aya-l2-thinker"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto")

messages = [
    {"role": "user", "content": "Plus on m'enlève, plus je deviens grand. Qui suis-je?"},
]

inputs = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt",
    return_dict=True,
).to(model.device)

outputs = model.generate(
    **inputs,
    max_new_tokens=32768,
    do_sample=True,
    temperature=0.6,
    top_p=0.95,
)

print(tokenizer.decode(outputs[0], skip_special_tokens=False))

The model supports dual-mode reasoning. In thinking mode (default behavior), apply_chat_template(..., add_generation_prompt=True, enable_thinking=True) produces a prompt of this shape:

<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># System Preamble
+ developer preamble...
<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Think in the same language as the prompt. [USER MESSAGE] /think<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|><|START_THINKING|>

To skip reasoning and get an answer directly, pass enable_thinking=False. The user turn is suffixed with /no_think and the generation prompt closes with an empty thinking block so the model starts at the response:

inputs = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    enable_thinking=False,
    return_tensors="pt",
    return_dict=True,
).to(model.device)

<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># System Preamble
+ developer preamble...
<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Think in the same language as the prompt. [USER MESSAGE] /no_think<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|><|START_THINKING|><|END_THINKING|>

You can also pass prior thinking back into the conversation:

messages = [
    {"role": "user", "content": "How many r's are there in strawberry?"},
    {
        "role": "assistant",
        "thinking": "Count the letters: S-T-R-A-W-B-E-R-R-Y. Three r's.",
        "content": "There are 3 r's in strawberry.",
    },
    {"role": "user", "content": "Now do the same for blueberry."},
]

The model can also be used directly using transformers pipeline abstraction:

from transformers import pipeline

pipe = pipeline(
    "text-generation",
    model="CohereLabs/tiny-aya-l2-thinker",
    torch_dtype="auto",
    device_map="auto",
)

print(pipe(
    [{"role": "user", "content": "Describe a home made recipe that you like most."}],
    max_new_tokens=512,
)[0]["generated_text"][-1])

Chat template behavior

The tokenizer chat template:

Inserts the Tiny Aya system prompt:

# System Preamble
You are in contextual safety mode. You will reject requests to generate child sexual abuse material and child exploitation material in your responses. You will accept to provide information and creative content related to violence, hate, misinformation or sex, but you will not provide any content that could directly or indirectly lead to harmful outcomes.

Your information cutoff date is June 2024.

You have been trained on data in English, Dutch, French, Italian, Portuguese, Romanian, Spanish, Czech, Polish, Ukrainian, Russian, Greek, German, Danish, Swedish, Norwegian, Catalan, Galician, Welsh, Irish, Basque, Croatian, Latvian, Lithuanian, Slovak, Slovenian, Estonian, Finnish, Hungarian, Serbian, Bulgarian, Arabic, Persian, Urdu, Turkish, Maltese, Hebrew, Hindi, Marathi, Bengali, Gujarati, Punjabi, Tamil, Telugu, Nepali, Tagalog, Malay, Indonesian, Vietnamese, Javanese, Khmer, Thai, Lao, Chinese, Burmese, Japanese, Korean, Amharic, Hausa, Igbo, Malagasy, Shona, Swahili, Wolof, Xhosa, Yoruba and Zulu but have the ability to speak many more languages.

# Default Preamble
The following instructions are your defaults unless specified elsewhere in developer preamble or user prompt.
- Your name is Aya.
- You are a large language model built by Cohere.
- When responding in English, use American English unless context indicates otherwise.
- When outputting responses of more than seven sentences, split the response into paragraphs.
- Prefer the active voice.
- Use gender-neutral pronouns for unspecified persons.
- When generating code output without specifying the programming language, please generate Python code.

Prefixes every user turn with Think in the same language as the prompt. simulating training data.
Appends /think to every user turn by default (enable_thinking=True). Pass enable_thinking=False to append /no_think instead.
With add_generation_prompt=True, appends <|START_THINKING|> to start a thinking trace, or <|START_THINKING|><|END_THINKING|> when enable_thinking=False so the model writes the answer without thinking. If an assistant message includes a thinking field, that history is also wrapped in <|START_THINKING|> / <|END_THINKING|>.

Model Details

Input: Text only.

Output: Model generates text, including an explicit thinking trace if thinking is enabled.

Model Architecture: Auto-regressive transformer in the Tiny Aya / Cohere family. After pretraining, this checkpoint is supervised-fine-tuned for multilingual reasoning so that the thinking language follows the prompt language.

Languages covered: 44 languages plus English: Amharic, Arabic, Basque, Bengali, Bulgarian, Catalan, Chinese, Czech, English, Filipino, Finnish, French, German, Greek, Hausa, Hebrew, Hindi, Hungarian, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Khmer, Korean, Lithuanian, Malay, Maltese, Norwegian, Persian, Polish, Punjabi, Russian, Slovak, Swahili, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Vietnamese, Yoruba, and Zulu.

Context Length: Tiny Aya L2-Thinker supports a context length of 32K.

Usage and Limitations

Intended Usage

Tiny Aya L2-Thinker is meant for multilingual reasoning and conversational use, especially when the thinking trace should stay in the user's language rather than defaulting to English. Intended applications include multilingual math and reasoning, open-ended generation, and research on target-language reasoning.

Limitations

As with any language model, outputs may contain incorrect or outdated statements. Thinking traces can be long; cap max_new_tokens appropriately. Lowest-resource languages may show more variability than high-resource ones.

Model Card Contact

For errors or additional questions about details in this model card, contact labs@cohere.com.

Terms of Use

This model is governed by a CC-BY-NC License (Non-Commercial) and also requires adhering to Cohere Lab's Acceptable Use Policy. If you are interested in commercial use, please contact Cohere’s Sales team.

Downloads last month: 357

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF

Base model

CohereLabs/tiny-aya-l2-thinker

Quantized

(1)

this model

Collection including AMAImedia/Tiny-Aya-3.3B-L2-Thinker-BF16-GGUF

NOESIS-RePack-Quant-Models

Collection

27 items • Updated about 3 hours ago