We introduce EXAONE Deep, which exhibits superior capabilities in various reasoning tasks including math and coding benchmarks, ranging from 2.4B to 32B parameters developed and released by LG AI Research. Evaluation results show that 1) EXAONE Deep 2.4B outperforms other models of comparable size, 2) EXAONE Deep 7.8B outperforms not only open-weight models of comparable scale but also a proprietary reasoning model OpenAI o1-mini, and 3) EXAONE Deep 32B demonstrates competitive performance against leading open-weight models.
This repository contains the AWQ-quantized weights of the reasoning 32B language model with the following features:
Number of Parameters (without embeddings): 30.95B
Number of Layers: 64
Number of Attention Heads: GQA with 40 Q-heads and 8 KV-heads
Vocab Size: 102,400
Context Length: 32,768 tokens
Quantization: AWQ with 4-bit group-wise weight-only quantization (W4A16g128)
Quickstart
We recommend to use transformers>=4.43.1 and autoawq>=0.2.8
Here is the code snippet to run conversational inference with the model:
python
1import torch
2from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
3from threading import Thread
45model_name ="LGAI-EXAONE/EXAONE-Deep-32B-AWQ"6streaming =True# choose the streaming option78model = AutoModelForCausalLM.from_pretrained(9 model_name,10 torch_dtype=torch.bfloat16,11 trust_remote_code=True,12 device_map="auto"13)14tokenizer = AutoTokenizer.from_pretrained(model_name)1516# Choose your prompt:17# Math example (AIME 2024)18prompt =r"""Let $x,y$ and $z$ be positive real numbers that satisfy the following system of equations:
19\[\log_2\left({x \over yz}\right) = {1 \over 2}\]\[\log_2\left({y \over xz}\right) = {1 \over 3}\]\[\log_2\left({z \over xy}\right) = {1 \over 4}\]
20Then the value of $\left|\log_2(x^4y^3z^2)\right|$ is $\tfrac{m}{n}$ where $m$ and $n$ are relatively prime positive integers. Find $m+n$.
2122Please reason step by step, and put your final answer within \boxed{}."""23# Korean MCQA example (CSAT Math 2025)24prompt =r"""Question : $a_1 = 2$인 수열 $\{a_n\}$과 $b_1 = 2$인 등차수열 $\{b_n\}$이 모든 자연수 $n$에 대하여\[\sum_{k=1}^{n} \frac{a_k}{b_{k+1}} = \frac{1}{2} n^2\]을 만족시킬 때, $\sum_{k=1}^{5} a_k$의 값을 구하여라.
2526Options :
27A) 120
28B) 125
29C) 130
30D) 135
31E) 140
3233Please reason step by step, and you should write the correct option alphabet (A, B, C, D or E) within \\boxed{}."""3435messages =[36{"role":"user","content": prompt}37]38input_ids = tokenizer.apply_chat_template(39 messages,40 tokenize=True,41 add_generation_prompt=True,42 return_tensors="pt"43)4445if streaming:46 streamer = TextIteratorStreamer(tokenizer)47 thread = Thread(target=model.generate, kwargs=dict(48 input_ids=input_ids.to("cuda"),49 eos_token_id=tokenizer.eos_token_id,50 max_new_tokens=32768,51 do_sample=True,52 temperature=0.6,53 top_p=0.95,54 streamer=streamer
55))56 thread.start()5758for text in streamer:59print(text, end="", flush=True)60else:61 output = model.generate(62 input_ids.to("cuda"),63 eos_token_id=tokenizer.eos_token_id,64 max_new_tokens=32768,65 do_sample=True,66 temperature=0.6,67 top_p=0.95,68)69print(tokenizer.decode(output[0]))
Note
The EXAONE Deep models are trained with an optimized configuration,
so we recommend following the Usage Guideline section to achieve optimal performance.
Evaluation
You can check the evaluation results of original EXAONE Deep models at GitHub or our documentation.
Deployment
EXAONE Deep models can be inferred in the various frameworks, such as:
TensorRT-LLM
vLLM
SGLang
llama.cpp
Ollama
LM-Studio
Please refer to our EXAONE Deep GitHub for more details about the inference frameworks.
Quantization
We provide the pre-quantized EXAONE Deep models with AWQ and several quantization types in GGUF format. Please refer to our EXAONE Deep collection to find corresponding quantized models.
Usage Guideline
To achieve the expected performance, we recommend using the following configurations:
Ensure the model starts with <thought>\n for reasoning steps. The model's output quality may be degraded when you omit it. You can easily apply this feature by using tokenizer.apply_chat_template() with add_generation_prompt=True. Please check the example code on Quickstart section.
The reasoning steps of EXAONE Deep models enclosed by <thought>\n...\n</thought> usually have lots of tokens, so previous reasoning steps may be necessary to be removed in multi-turn situation. The provided tokenizer handles this automatically.
Avoid using system prompt, and build the instruction on the user prompt.
Additional instructions help the models reason more deeply, so that the models generate better output.
For math problems, the instructions "Please reason step by step, and put your final answer within \boxed{}." are helpful.
For more information on our evaluation setting including prompts, please refer to our Documentation.
In our evaluation, we use temperature=0.6 and top_p=0.95 for generation.
When evaluating the models, it is recommended to test multiple times to assess the expected performance accurately.
Limitation
The EXAONE language model has certain limitations and may occasionally generate inappropriate responses. The language model generates responses based on the output probability of tokens, and it is determined during learning from training data. While we have made every effort to exclude personal, harmful, and biased information from the training data, some problematic content may still be included, potentially leading to undesirable responses. Please note that the text generated by EXAONE language model does not reflects the views of LG AI Research.
Inappropriate answers may be generated, which contain personal, harmful or other inappropriate information.
Biased responses may be generated, which are associated with age, gender, race, and so on.
The generated responses rely heavily on statistics from the training data, which can result in the generation of
semantically or syntactically incorrect sentences.
Since the model does not reflect the latest information, the responses may be false or contradictory.
LG AI Research strives to reduce potential risks that may arise from EXAONE language models. Users are not allowed
to engage in any malicious activities (e.g., keying in illegal information) that may induce the creation of inappropriate
outputs violating LG AI’s ethical principles when using EXAONE language models.
