fluently-lm/FluentlyLM-Prinum

FluentlyLM Prinum (32B-version)

Introducing the first standalone model from Project Fluently LM! We worked on it for several months, used different approaches, and eventually found the optimal one.

Model Details

Model Description

• Developed by: @fluently-lm
• Model type: Causal Language Models (QwenForCausalLM, LM Transformer)
• Number of Parameters: 32.5B
• Number of Paramaters (Non-Embedding): 31.0B
• Number of Layers: 64
• Number of Attention Heads (GQA): 40 for Q and 8 for KV
• Context Length: Full 131,072 tokens
• Language(s) (NLP): English, French, Spanish, Russian, Chinese, Japanese, Persian (official support)
• License: MIT

Quickstart

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "fluently-lm/FluentlyLM-Prinum"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "Write a quick sort algorithm."
messages = [
    {"role": "system", "content": "You are FluentlyLM, created by Project Fluently. You are a helpful assistant."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=1024
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

GGUF-using

You can also use our model locally via GGUF file in various interfaces and workflows, we offer several repos for downloading GGUF:

• mradermacher/FluentlyLM-Prinum-GGUF (all GGUF-quants)
• fluently-lm/FluentlyLM-Prinum-Q4_K_M-GGUF (only Q4_K_M-quant) (coming soon...)

Model recipe

Evolution

🏆 12th place on Open LLM Leaderboard

Special thanks

🤗 We are grateful for open source resources, technologies and assistance from: Unsloth AI, Axolotl AI, Argilla, Alibaba Cloud: Qwen, NVIDIA and NousResearch.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	47.22
IFEval (0-Shot)	80.90
BBH (3-Shot)	59.48
MATH Lvl 5 (4-Shot)	54.00
GPQA (0-shot)	18.23
MuSR (0-shot)	17.26
MMLU-PRO (5-shot)	53.42