Lythri

Hugging Face | ModelScope | Technical Report

Introduction

Lythri-Wink-2B is part of the Lythri 2.0 family, with 2.52B total parameters and 1.48B activated parameters. It is a compact, direct-response model optimized for on-device conversation and emotional support.

Compared with the larger Lythri-Tada-4B, Wink focuses on three practical features:

  • Smaller deployment footprint. Wink substantially reduces both total and activated parameters.
  • Multi-turn conversations. Wink is trained on multi-turn companion conversations and earlier-turn context.
  • System prompt support. Like Lythri-Tada-4B, Wink follows system prompts, so you can set a persona, tone or rules for your app directly in the system message.

Wink trades knowledge, reasoning, math and coding performance for its smaller deployment footprint; it is not intended to match the larger Lythri-Tada-4B on general benchmarks.

Writing Naturalness

EQ-Bench Slop Score

Within the same family, Lythri-Tada-4B scores 39.0 on the EQ-Bench Slop Score and writes more naturally than GPT-4o (47.7) and DeepSeek-R1 (67.9), far ahead of Qwen3-4B (62.2) and Gemma-3-4B (85.2).

General Benchmarks

Benchmark Lythri-Tada-4B Lythri-Wink-2B
Knowledge
MMLU55.4139.32
ARC-C56.5741.13
Reasoning
HellaSwag74.2358.65
General
SocialIQA51.4345.24
Science
GPQA Diamond21.2123.23
Math
GSM8K27.4513.04
Code / Instruction
HumanEval24.397.93
IFEval39.1923.48

The listed benchmarks use their standard task settings and few-shot counts with the chat template applied. Lythri-Wink-2B is evaluated as a Q8_0 GGUF.

Quickstart

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_path = "Lythri/Lythri-Wink-2B"
tok = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(
    model_path, dtype=torch.bfloat16, device_map="auto"
)

messages = [{"role": "user", "content": "My friend just lost their job and seems really down. What should I say to them?"}]
chat = tok.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tok(chat, return_tensors="pt", add_special_tokens=False).to(model.device)

with torch.inference_mode():
    out = model.generate(
        **inputs,
        max_new_tokens=1024,
        do_sample=True,
        temperature=0.8,
        top_p=0.95,
        eos_token_id=[1, 106],
    )

print(tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

Recommended Generation Config

generation_config = {
    "temperature": 0.8,
    "top_p": 0.95,
    "top_k": 64,
    "max_new_tokens": 1024,
    "repetition_penalty": 1.05,
    "do_sample": True,
    "eos_token_id": [1, 106],
}

out = model.generate(**inputs, **generation_config)

System Prompt (Optional)

Lythri-Wink-2B supports system prompts. It works well without one; to set a persona or rules for your app, add a system message:

messages = [
    {"role": "system", "content": "You are Lythri, a warm and caring companion. Reply in the same language as the user."},
    {"role": "user", "content": "I had a really rough day at work."},
]

Compute

Lythri-Wink-2B is structurally sheared from the Gemma4-E2B PT model and recovered through teacher distillation. It is then trained with CoT SFT, followed by instant-reply SFT with multi-turn companion conversations and a small writing-only online preference-optimization stage judged by Gemma2-27B.

Limitations

  • Lythri is optimized for conversation and emotional understanding, not for math, coding or complex reasoning.
  • Lythri is not a substitute for professional mental health support. If you or someone you know is in crisis, please contact local emergency services or a crisis helpline.
  • Like all language models, it can produce inaccurate or inappropriate content.

License

Lythri is built on Gemma 4 and is released under the Apache License 2.0.

Citation

@techreport{li2026lythri,
  title       = {Lythri Technical Report},
  author      = {Li, Jiawen},
  year        = {2026},
  institution = {Zenodo},
  doi         = {10.5281/zenodo.23179311},
  url         = {https://doi.org/10.5281/zenodo.23179311}
}

Support

The whole training process is self-funded. If you like our model, please click a free like — that means a lot to me as a student!

Downloads last month
-
Safetensors
Model size
3B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Lythri/Lythri-Wink-2B

Finetuned
(162)
this model
Quantizations
3 models

Collection including Lythri/Lythri-Wink-2B