Senti is a specialized LLM for cryptocurrency market sentiment analysis. This is the SFT (Supervised Fine-Tuning) base model, trained on crypto summaries, sentiment classification, and chat data.

Training Pipeline

Qwen/Qwen3-8B (Base, 8B params)
       │
       ▼
┌─────────────────────────────────────┐
│         SFT Training                │
│   9,314 samples (train + val)       │
│   5.34 hours on A100 80GB           │
└─────────────────────────────────────┘
       │
       ▼
   This Model ────► DPO Training ────► senti-qwen3-8b-dpo (recommended)

Model Details

Property	Value
Base Model	Qwen/Qwen3-8B
Parameters	8 Billion (Dense)
Languages	English (80%), Turkish (20%)
License	Apache 2.0
Fine-tuning Method	QLoRA (4-bit NF4)

Training Data

Overview

Dataset	Samples	Size	Purpose
Training Set	8,383	22.4 MB	Supervised fine-tuning
Validation Set	931	2.4 MB	Evaluation
Total	9,314	24.8 MB

Task Distribution

Task	Samples	Percentage	Description
Summary Generation	4,491	53.6%	Generate market summaries from tweets
Sentiment Classification	3,624	43.2%	Classify individual tweet sentiment
Chat / Q&A	268	3.2%	Answer crypto market questions

Summary Task Breakdown

Summary Type	Count	Description
Hourly Summaries	4,086	Real-time market pulse from recent tweets
Daily Summaries	405	End-of-day market recap and synthesis

Tweets Per Summary

Metric	Value
Average	44.4 tweets
Minimum	5 tweets
Maximum	100 tweets
Total Tweets Processed	~199,000

Input/Output Statistics

Metric	Characters	~Tokens
Avg Input Length	2,167	~540
Avg Output Length	995	~250
Max Sequence Length	4,096	4,096

Sentiment Classification Details

5-Class System:

Label	Description	Example
`STRONG_BULLISH`	Extreme optimism, price targets, euphoria	"BTC to 200k! This is just the beginning!"
`BULLISH`	Positive outlook, accumulation mentions	"Adding more ETH here, looking good for Q1"
`NEUTRAL`	Factual, news, no clear sentiment	"Bitcoin trading at $97,500 today"
`BEARISH`	Concern, caution, negative outlook	"Not liking this price action, reducing exposure"
`STRONG_BEARISH`	Panic, crash predictions, extreme fear	"This is the top, selling everything"

Classification Distribution (3,624 samples):

Label	Count	Percentage
BULLISH	~1,200	33.1%
NEUTRAL	~1,100	30.3%
STRONG_BULLISH	~650	17.9%
BEARISH	~450	12.4%
STRONG_BEARISH	~224	6.2%

Coins Covered

The model was trained on data from 12 cryptocurrencies:

Coin	Ticker	Category	Sample Distribution
Bitcoin	BTC	Major	~18%
Ethereum	ETH	Major	~16%
Solana	SOL	Layer 1	~12%
XRP	XRP	Payment	~10%
BNB Chain	BNB	Layer 1	~9%
Chainlink	LINK	Oracle	~8%
Arbitrum	ARB	Layer 2	~7%
Avalanche	AVAX	Layer 1	~6%
Hyperliquid	HYPE	DeFi	~5%
0G	0G	AI/Infra	~4%
Aster	ASTER	Emerging	~3%
Zcash	ZEC	Privacy	~2%

Data Sources

Source	Description
Twitter/X	Real-time tweets from crypto KOLs and communities
Time Range	October 2025 - January 2026
Languages	English (primary), Turkish (secondary)
Filtering	Relevance scoring, spam removal, deduplication

Training Configuration

SFT Training Parameters

Parameter	Value
Base Model	Qwen/Qwen3-8B
Method	QLoRA (4-bit NF4)
Epochs	3
Batch Size	16 (4 × 4 grad accum)
Learning Rate	1e-4
Warmup Ratio	0.05
LR Scheduler	Cosine
Weight Decay	0.01
Max Sequence Length	4,096 tokens
Training Time	5.34 hours
Hardware	NVIDIA A100 80GB

LoRA Configuration

Parameter	Value
Rank (R)	32
Alpha	64
Dropout	0.05
Target Modules	q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj

Quantization (QLoRA)

Parameter	Value
Bits	4-bit
Quant Type	NF4
Compute Dtype	bfloat16
Double Quant	Yes

Model Capabilities

1. Summary Generation

Input: Raw tweets about a cryptocurrency Output: Structured market analysis

markdown

## BTC Hourly Summary

**Overall Sentiment: BULLISH (7.2/10)**

### Key Developments
- Bitcoin testing $98,000 resistance level
- ETF inflows remain positive
- Whale activity detected on-chain

### Risk Factors
- Some profit-taking at round numbers
- RSI elevated but not overbought

### Outlook
Momentum positive. Watch $100K level.

2. Sentiment Classification

Input: Single tweet Output: Sentiment label

Input: "ETH looking strong here, accumulating more"
Output: BULLISH

Supported Labels:

STRONG_BULLISH
BULLISH
NEUTRAL
BEARISH
STRONG_BEARISH

3. Chat / Q&A

Input: Question about crypto market Output: Informative answer

Q: "What's the current sentiment on SOL?"
A: "Based on recent social media activity, Solana sentiment is moderately
    bullish. Key themes include ecosystem growth, DeFi TVL increases, and
    positive price action. Some concerns exist about network stability
    during high-traffic periods."

Usage

Basic Usage

python

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load model
base_model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen3-8B",
    torch_dtype=torch.bfloat16,
    device_map="auto"
)
model = PeftModel.from_pretrained(base_model, "dorukardahan/senti-qwen3-8b-sft")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")

# Generate
messages = [
    {"role": "system", "content": "You are Senti, a crypto sentiment analyst."},
    {"role": "user", "content": "Summarize BTC market sentiment."}
]

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

outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.7)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

With 4-bit Quantization (Lower VRAM)

python

from transformers import BitsAndBytesConfig

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
)

base_model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen3-8B",
    quantization_config=bnb_config,
    device_map="auto"
)
model = PeftModel.from_pretrained(base_model, "dorukardahan/senti-qwen3-8b-sft")

Deployment Options

Format	Size	Platform	Latency	Memory
FP16 (native)	~16 GB	GPU	~1-2s	20+ GB VRAM
INT8 / Q8_0	~8 GB	GPU/CPU	~2-5s	12+ GB
INT4 / Q4_K_M	~4.5 GB	CPU	~5-10s	8+ GB RAM

Limitations

This is the SFT base model; DPO version is recommended for production
Trained primarily on English (80%) and Turkish (20%) data
Knowledge reflects training data up to January 2026
Optimized for 12 specific cryptocurrencies
Should not be used as sole basis for financial decisions
May reflect biases present in social media data

Intended Use

Cryptocurrency market sentiment analysis
Social media summarization for research
Base for further preference optimization (DPO/RLHF)
Experimentation and comparison with DPO version

Not Intended For:

Financial advice
Automated trading decisions without human oversight
Production deployment (use DPO version instead)

Related Models

Model	Purpose	Link
DPO Optimized (Recommended)	Production-ready	dorukardahan/senti-qwen3-8b-dpo

Citation

bibtex

@misc{senti-qwen3-8b-sft,
  author = {Doruk Ardahan},
  title = {Senti: Crypto Sentiment Analysis Model (SFT)},
  year = {2025},
  publisher = {HuggingFace},
  url = {https://huggingface.co/dorukardahan/senti-qwen3-8b-sft}
}

Acknowledgments

Base Model: Qwen Team for Qwen3-8B
Training Framework: HuggingFace TRL
Infrastructure: RunPod for A100 GPU access