---
language:
- en
license: apache-2.0
pipeline_tag: text-generation
tags:
- x
- twitter
- tweet
- persona
- router
- lora
- peft
- qlora
base_model:
- Qwen/Qwen3-14B
- google/gemma-2-9b-it
- meta-llama/Meta-Llama-3.1-8B-Instruct
- microsoft/Phi-3.5-mini-instruct
---

# Miko X Tweet Ensemble — multi-base, router-driven LoRA stack

> This model has been trained using [Miko](https://x.com/project_miko), the fully autonomous AI agent for [Miko Protocol](https://mikoprotocol.com).

[![Miko on X](https://img.shields.io/badge/Miko-%40project_miko-blue?logo=x)](https://x.com/project_miko) [![Miko Protocol Website](https://img.shields.io/badge/Website-Miko%20Protocol-lightgrey)](https://mikoprotocol.com)

<img src="https://ik.imagekit.io/ueitnjew7/miko_huggingface.png">

## What it is
Miko is a multi-base, multi-adapter ensemble **built for X/Twitter**.
It discovers style clusters from real tweets, fine-tunes **one LoRA per style**, and routes your prompt to the best-fit style at runtime.

---

## Why it’s different
- **Multi-base adapters by design.**
  Not tied to a single model family. Style adapters originate from multiple bases:
- `Qwen/Qwen3-14B`
- `google/gemma-2-9b-it`
- `meta-llama/Meta-Llama-3.1-8B-Instruct`
- `microsoft/Phi-3.5-mini-instruct`

- **X-native behavior.**
  Short form, emoji/hashtag cadence, memes/irony, and fast “CT” tone.

- **Router that understands styles.**
  Uses Qwen3-14B hidden states with prototype similarity + a small projection head to pick a style before generation.

---

## Base models & typical roles (observed tendencies)
| Base model | Typical role / personality | Good for |
|---|---|---|
| **Qwen/Qwen3-14B** | Router backbone & **fallback generator**. Balanced, hashtag-friendly. | General comments, quick Q/A, mentions |
| **mistralai/Mistral-Nemo-Instruct-2407** | **Crisp technical tone**, list-y facts, tight bullets. | Alpha/launch notes, “3-point” updates |
| **google/gemma-2-9b-it** | **Smooth and narrative**; softer, reflective voice. | Story-like replies, mini-threads |
| **meta-llama/Meta-Llama-3.1-8B-Instruct** | **Clear directives / neutral composition.** | How-to tweets, best practices |
| **microsoft/Phi-3.5-mini-instruct** | **Snappy one-liners**; memes/emoji friendly. | Witty hooks, irony, punchy replies |

*(Roles are tendencies learned from tweet data; they’re not hard rules.)*

## Training data

**Proprietary — Miko Agent Tweet Corpus.**
Tweets authored by the fully-autonomous X (Twitter) agent **[Miko(@project_miko)](https://x.com/project_miko)**, collected from the live account’s public timeline and agent logs under the account owner's control.
– Domain: Crypto/X discourse (emojis, hashtags, memes, irony)
– Time window: rolling weekly refreshes (e.g., 7–14 days)
– Redistribution: the raw dataset is **not** redistributed; only model weights are shared.
(Preprocessing: light normalization/filters, deduplication; style clustering via HDBSCAN.)

---

## How it works (high-level)
1. **Style discovery** — cluster tweet embeddings (e.g., HDBSCAN) to assign **style IDs**.
2. **Per-style LoRA** — train one adapter per style, possibly from different base models.
3. **Routing** — Qwen3-14B features → prototype similarity + projection head → pick a style.
4. **Generation** — load the chosen base, attach the matching LoRA, generate with a light `<style_{id}>` tag.

---

## Quickstart
```python
from inference import MikoEnsemble

ens = MikoEnsemble(".")
print(ens.generate("CT keeps fading this rally. What's your take?"))
```

## Force a style (advanced)
```python
def generate_with_style(ens, sid, prompt, **gen):
    styled = f"<style_{sid}>{prompt}"
    model, tok = ens._load_adapter_with_base(sid)
    ipt = tok(styled, return_tensors="pt", truncation=True, max_length=256, padding=True).to(model.device)
    out = model.generate(
        **ipt,
        max_new_tokens=gen.get("max_new_tokens", 120),
        temperature=gen.get("temperature", 0.8),
        do_sample=True,
        top_p=0.95,
        pad_token_id=tok.pad_token_id,
        eos_token_id=tok.eos_token_id,
    )
    return tok.decode(out[0], skip_special_tokens=True).replace(styled, "").strip()
```

## VRAM & speed tips
- 4-bit (nf4, double-quant, bf16 compute) supported; 16–24GB VRAM is enough for one adapter at a time.
- A small LRU cache keeps recently used styles in memory (default 2).

---

## Files
- `lora_adapters/style_{id}_lora/` — per-style LoRA folder (with its adapter_config.json).
- `router/router_state.pt` — router head (prototypes + projection).
- `inference.py` — lazy loader + generator.
- `README_METADATA.json` — style IDs, number of styles, base list, timestamp.

---

## Intended use (tweet personas)
- Witty/ironic one-liners — hooks, memes, playful replies
- Tech/alpha notes — launch takeaways, bullet summaries, link threads
- Narrative reframing — bullish/bearish angles, story-style posts
- Q&A / reply bots — short, clear responses in mentions/threads

## Limitation
- Optimized for tweets & short threads; not a general chatbot.
- Each base model retains its own license/terms.

## License
Apache-2.0.

## Acknowledgements
Thanks to the Qwen, Mistral, Gemma-2, Llama-3.1, and Phi-3.5 communities.


## Changelog
- 2026-01-03: weekly refresh (days=7); retrained adapters & router.