Spaces:
Running
Running
File size: 14,735 Bytes
b503124 f0c89ea b503124 9e8a37f f0c89ea 9e8a37f f0c89ea 9e8a37f 64e4100 9e8a37f a61cf69 9e8a37f e2edb82 64e4100 9e8a37f 64e4100 9e8a37f 64e4100 9e8a37f f0c89ea 9e8a37f f0c89ea 64e4100 9e8a37f 64e4100 9e8a37f 71bf51a ae12dba dedade8 3fb94b5 dedade8 a93374f cb0f440 a93374f cb0f440 9204c81 d34f468 9204c81 8880537 cb0f440 8880537 34abae0 71bf51a 34abae0 0f159dc 3fb94b5 0fcf800 14037f6 01220d9 14037f6 2248f68 d34f468 2248f68 d2f0ab5 9e8a37f d2f0ab5 d34f468 d2f0ab5 9e8a37f f0c89ea c7886eb 9e8a37f f0c89ea 64e4100 9e8a37f 8880537 34abae0 0f159dc 9e8a37f f0c89ea 9e8a37f 64e4100 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 | ---
title: README
emoji: πΈ
colorFrom: indigo
colorTo: purple
sdk: static
pinned: false
---
<div align="center">
# πΈ agentic-underground
**Open tools and research for building software with AI agents** β
disciplined workflows, safety, token efficiency, and reproducible model research.
<a href="https://huggingface.co/collections/agentic-underground/prattle-not-6af7ac5e95e0fc80de563d61"><img alt="Collection: prattle_not" src="https://img.shields.io/badge/collection-prattle__not-6366f1"></a>
<a href="https://agentic-underground.github.io/prattle_not/"><img alt="Live demo in your browser" src="https://img.shields.io/badge/demo-runs%20in%20your%20browser-a855f7"></a>
<a href="https://github.com/agentic-underground"><img alt="Source on GitHub" src="https://img.shields.io/badge/source-GitHub-24292f?logo=github"></a>
<a href="https://huggingface.co/papers/2609.25053"><img alt="Builds on LatentPort by Simon Villani (arXiv 2609.25053)" src="https://img.shields.io/badge/builds%20on-LatentPort%20by%20Simon%20Villani-b31b1b"></a>
</div>
---
> **β οΈ Security note for every bridge here.** A bridged model does not read the prompt itself, the system
> prompt included, and rules set in a system prompt are broken more often bridged (measured on six pairs).
> Do not rely on system-prompt instructions as guardrails for a bridged model; enforce policy outside it.
## On the Hub: the prattle_not family
**prattle_not** reproduces [**LatentPort**](https://huggingface.co/papers/2609.25053), Simon Villani's method for
handing a conversation from one language model to another *through its memory*. A small model reads
the prompt; a fitted **bridge** translates its key/value cache into the memory a larger model would
have built; the larger model answers from it β **without reading the prompt itself**, so its first
token arrives sooner.
What we publish here are the **bridges**: the fitted weights that make that handoff work between a
specific pair of models. Everything is gathered in the
**[prattle_not collection β](https://huggingface.co/collections/agentic-underground/prattle-not-6af7ac5e95e0fc80de563d61)**
### β¨ Featured β [`smollm2-135m-to-360m-bridge`](https://huggingface.co/agentic-underground/smollm2-135m-to-360m-bridge)
SmolLM2-135M reads; SmolLM2-360M, almost three times larger, answers. Two models with different
shapes, bridged with the paper's **one-token handoff** and its learned correction.
| | |
|:--|:--|
| β‘ **Time to first token** | **1.87Γ** sooner (95% CI 1.83β1.91), in WebAssembly, on a CPU |
| πͺ **Bridge tokens** | **1** β the 360M reads one token of the prompt itself |
| π― **Held-out questions** | **6 of 7** correct β as many as the 360M reading everything |
| π **Fidelity (NCR)** | **0.930** β the paper's corrected bridge on its own pair: 0.918 |
βΆοΈ **[Try it in your browser](https://agentic-underground.github.io/prattle_not/)** β both models and the
bridge download once and run locally. Turn the bridge on, ask anything, and watch the counters show
who read what.
### π The reproduction behind it
Before carrying the method to new models, we reproduced LatentPort's own **4B β 27B** pair on two
RTX 3090s with upstream's timing code and statistic: **1.36Γ** sooner at 128 prompt tokens, rising to
**2.94Γ** at ~2K and **3.39Γ** at ~16K. The full tables, every run, and a step-by-step guide to
repeat them are in the [prattle_not repository](https://github.com/agentic-underground/prattle_not).
### π The paper's own pair β [`qwen3.5-4b-to-9b-bridge`](https://huggingface.co/agentic-underground/qwen3.5-4b-to-9b-bridge)
Qwen3.5-4B reads; Qwen3.5-9B answers. This is the pair Simon Villani's paper fits, re-fitted
independently with its method. The attention layers' keys and values are mapped, and the 24
recurrent layers cross as the 4B's own state.
| | |
|:--|:--|
| π **Fidelity (NCR)** | **0.989** with the paper's learned correction (0.974 from the maps alone), our highest yet |
| π§ͺ **Control (wrong donor)** | 0.308, high because the recurrent state carries general context whichever prompt made it |
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
| β‘ **Time to first token** | **1.16Γ** sooner (95% CI 1.16β1.16) on two RTX 3090s at ~256 tokens; ceiling 1.38Γ. Small: the two models share a recurrent core that costs the same in both |
### 𧬠Qwen3 at a larger size β [`qwen3-4b-to-8b-bridge`](https://huggingface.co/agentic-underground/qwen3-4b-to-8b-bridge)
Qwen3-4B reads; Qwen3-8B answers, across all 36 layers.
| | |
|:--|:--|
| π **Fidelity (NCR)** | **0.890** with the paper's learned correction |
| π§ͺ **Control (wrong donor)** | β0.026 |
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
| β‘ **Time to first token** | **1.12Γ** sooner (95% CI 1.12β1.12) on two RTX 3090s at ~256 tokens; ceiling 1.61Γ |
### π Fifty-four times larger β [`qwen3-0.6b-to-32b-bridge`](https://huggingface.co/agentic-underground/qwen3-0.6b-to-32b-bridge)
Qwen3-0.6B reads; Qwen3-32B, quantized to Q4K, answers. Published with its miss stated.
| | |
|:--|:--|
| β‘ **Time to first token** | **4.21Γ** sooner (95% CI 4.19β4.23) at ~256 tokens, **7.17Γ** at ~1K, on two RTX 3090s; ceilings 11.60Γ and 14.33Γ |
| π― **Held-out questions** | **6 of 7** β the 32B alone gets 7, so it is **below our answer bar** |
| π **Fidelity (NCR)** | **0.866** with the paper's learned correction (0.794 from the maps alone) |
| π§ͺ **Control (wrong donor)** | 0.197 |
### πͺ A 13.6Γ gap, at the bar β [`qwen3-0.6b-to-8b-bridge`](https://huggingface.co/agentic-underground/qwen3-0.6b-to-8b-bridge)
Qwen3-0.6B reads; Qwen3-8B, 13.6 times larger, answers.
| | |
|:--|:--|
| β‘ **Time to first token** | **2.82Γ** sooner (95% CI 2.82β2.83) on one RTX 3090 at ~256 tokens; ceiling 6.03Γ |
| π― **Held-out questions** | **7 of 7**, the same as the 8B alone |
| π **Fidelity (NCR)** | **0.803** with the paper's learned correction |
| π§ͺ **Control (wrong donor)** | 0.092 |
### π A larger gap β [`smollm2-360m-to-1.7b-bridge`](https://huggingface.co/agentic-underground/smollm2-360m-to-1.7b-bridge)
SmolLM2-360M reads; SmolLM2-1.7B, about five times larger and shaped differently, answers from
**one bridge token**. Here the two models differ on the answer test, so it can show a gain.
| | |
|:--|:--|
| π― **Held-out questions** | **7 of 7**, as the 1.7B reading everything (the 360M alone: 6) |
| π **Fidelity (NCR)** | **0.845** with the paper's learned correction (0.611 from the maps alone) |
| π§ͺ **Control (wrong donor)** | β0.266 |
| β‘ **Time to first token** | **1.67Γ** sooner (95% CI 1.66β1.69) on one RTX 3090 at ~256 tokens; ceiling 2.59Γ |
### π A third family β [`gemma-3-270m-to-1b-bridge`](https://huggingface.co/agentic-underground/gemma-3-270m-to-1b-bridge)
Gemma 3 270M reads; Gemma 3 1B answers, across sliding-window and global attention layers.
| | |
|:--|:--|
| β‘ **Time to first token** | **1.91Γ** sooner (95% CI 1.86β1.97) on one RTX 3090 at ~256 tokens; ceiling 2.79Γ |
| π **Fidelity (NCR)** | **0.870**, but a wrong prompt's memory reaches 0.600 here: an empty Gemma 3 1B is far off, so read the two together |
| π― **Held-out questions** | **6 of 7**, the same as the 270M alone (the 1B: 7) β no answer gain yet |
βοΈ **Gated, and not Apache-2.0:** it is a Gemma Model Derivative, distributed under the
[Gemma Terms of Use](https://ai.google.dev/gemma/terms).
### π¦ A fourth family β [`llama-3.2-1b-to-3b-bridge`](https://huggingface.co/agentic-underground/llama-3.2-1b-to-3b-bridge)
**Built with Llama.** Llama 3.2 1B reads; Llama 3.2 3B answers, across different head widths (64
and 128).
| | |
|:--|:--|
| π **Fidelity (NCR)** | **0.965** with the paper's learned correction (0.818 from the maps alone) |
| π§ͺ **Control (wrong donor)** | β0.040 |
| π― **Held-out questions** | **6 of 7**, the same as the 3B alone (the 1B: 7). No answer gain yet |
| β‘ **Time to first token** | **1.70Γ** sooner (95% CI 1.66β1.74) on one RTX 3090 at ~256 tokens; ceiling 2.54Γ |
βοΈ **Gated, and not Apache-2.0:** it is a derivative work of Llama 3.2, distributed under the
[Llama 3.2 Community License](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct/blob/main/LICENSE.txt).
### π Short convolutions β [`lfm2-350m-to-1.2b-bridge`](https://huggingface.co/agentic-underground/lfm2-350m-to-1.2b-bridge)
LFM2-350M reads; LFM2-1.2B answers. Liquid AI's LFM2 mixes short convolutions with attention, and
the two sizes differ in width, so the convolution state cannot be copied: the bridge maps it too.
| | |
|:--|:--|
| β‘ **Time to first token** | **1.88Γ** sooner (95% CI 1.86β1.91) on one RTX 3090 at ~256 tokens; ceiling 2.50Γ |
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
| π **Fidelity (NCR)** | **0.978** (0.982 from the maps alone), but a wrong prompt's memory reaches 0.730: an empty LFM2 is far off, so read the two together |
βοΈ **Gated, and not Apache-2.0:** it is distributed under the base models'
[LFM Open License v1.0](https://huggingface.co/agentic-underground/lfm2-350m-to-1.2b-bridge/blob/main/LICENSE),
which does not license Commercial Use by an entity with USD 10M or more in annual revenue.
### π Ninety-five times larger, through Mamba-2 β [`granite-4.0-h-350m-to-h-small-bridge`](https://huggingface.co/agentic-underground/granite-4.0-h-350m-to-h-small-bridge)
Granite 4.0 H 350M reads; H Small, a 72-expert 32B quantized to Q4K, answers from translated Mamba-2
inputs. The paper's correction scores higher but answers 6 of 7, so the uncorrected maps are
published.
| | |
|:--|:--|
| π― **Held-out questions** | **7 of 7**, the same as H Small alone (the 350M alone gets 5) |
| π **Fidelity (NCR)** | **0.767** from the maps alone |
| π§ͺ **Control (wrong donor)** | 0.161 |
| β‘ **Time to first token** | **1.75Γ** sooner (95% CI 1.74β1.75) on two RTX 3090s at ~256 tokens, **1.62Γ** at ~1K; ceilings 7.30Γ and 7.58Γ. H Small rebuilds 36 Mamba-2 states at the handoff |
### π Gemma 4, two layer shapes β [`gemma-4-e2b-to-31b-bridge`](https://huggingface.co/agentic-underground/gemma-4-e2b-to-31b-bridge)
Gemma 4 E2B reads; Gemma 4 31B, quantized to Q4K, answers. Its 50 sliding layers (16 Γ 256) and 10
global layers (4 Γ 512) are each fitted in their own shape. Apache-2.0, not gated.
| | |
|:--|:--|
| π― **Held-out questions** | **7 of 7**, the same as the 31B alone |
| π **Fidelity (NCR)** | **0.809** from the maps alone |
| π§ͺ **Control (wrong donor)** | 0.448 |
| β‘ **Time to first token** | **1.97Γ** sooner (95% CI 1.94β2.00) on two RTX 3090s at ~256 tokens; ceiling 5.10Γ. Longer prompts await an attention path that fits a 24 GB card |
### π Mamba-2, translated β [`granite-4.0-h-1b-to-h-tiny-bridge`](https://huggingface.co/agentic-underground/granite-4.0-h-1b-to-h-tiny-bridge)
Granite 4.0 H 1B reads; H Tiny, a 64-expert mixture, answers. Mamba-2 state cannot be copied between
separately trained sizes, so the bridge translates each recurrent layer's **inputs**, and H Tiny
builds its own state from them.
| | |
|:--|:--|
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
| π **Fidelity (NCR)** | **0.924** with the paper's learned correction (0.916 from the maps alone) |
| π§ͺ **Control (wrong donor)** | β0.175 |
| β‘ **Time to first token** | **2.06Γ** sooner (95% CI 1.95β2.18) on one RTX 3090 at ~256 tokens; ceiling 3.95Γ. Its experts route on the host CPU, so a busy CPU lowers it |
### π A hybrid model β [`qwen3.5-0.8b-to-2b-bridge`](https://huggingface.co/agentic-underground/qwen3.5-0.8b-to-2b-bridge)
Qwen3.5-0.8B reads; Qwen3.5-2B answers. Qwen3.5 mixes full-attention layers with Gated DeltaNet
layers: the bridge maps the attention layers' keys and values, and the recurrent state crosses as
the 0.8B's own.
| | |
|:--|:--|
| π **Fidelity (NCR)** | **0.917** with the paper's learned correction (0.884 from the maps alone) |
| π§ͺ **Control (wrong donor)** | 0.019 |
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
| β‘ **Time to first token** | **1.36Γ** sooner (95% CI 1.35β1.36) on one RTX 3090 at ~256 tokens; ceiling 1.50Γ |
### 𧬠A second family β [`qwen3-0.6b-to-1.7b-bridge`](https://huggingface.co/agentic-underground/qwen3-0.6b-to-1.7b-bridge)
Qwen3-0.6B reads; Qwen3-1.7B answers. This is the real test of whether Simon Villani's method
carries beyond one family, fitted with the same tools and measured on held-out questions.
| | |
|:--|:--|
| π **Fidelity (NCR)** | **0.805** with the paper's learned correction (0.645 from the maps alone) |
| π― **Held-out questions** | **7 of 7** correct, as many as the 1.7B reading everything |
| π§ͺ **Control (wrong donor)** | β0.127: the bridge fed another question's memory scores far lower |
| β‘ **Time to first token** | **1.47Γ** sooner (95% CI 1.46β1.48) on one RTX 3090 at ~256 tokens; ceiling 2.03Γ |
Each new pair gets its own repository here and joins the collection once its weights are fitted
and measured.
---
## How we publish
- π¬ **Measured** β every number on a card says how it was measured, and the speed-ups can be
measured again on your own hardware with the published runtime.
- π **Pinned** β our pages fetch weights by commit, never by `main`, so a later upload cannot change
what a published result measured.
- π§© **Our own work only** β we host what we fitted. Base models stay with their publishers and are
linked, never mirrored.
- π **Attributed** β prattle_not is a derivative work of [LatentPort](https://latentport.com/) by
Simon Villani ([source](https://github.com/LatentPort/latentport), Apache-2.0). Our bridges are
Apache-2.0, except those fitted from Gemma, which are Gemma Model Derivatives, gated and distributed
under the [Gemma Terms of Use](https://ai.google.dev/gemma/terms), and those fitted from Llama, which
are Built with Llama, gated and distributed under the Llama 3.2 Community License, and those fitted
from LFM2, gated and distributed under the LFM Open License v1.0.
## Elsewhere
The **source** for everything here, and our other tools for building software with agents (Claude
Code plugins, a static auditor for agent skills, a token-aware scheduler), live on
**[GitHub β agentic-underground](https://github.com/agentic-underground)**.
Questions and corrections are welcome: open a discussion on the model concerned, or an
[issue on prattle_not](https://github.com/agentic-underground/prattle_not/issues).
|