Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- README.md +51 -0
- slipstream-Q8_0.gguf +3 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
slipstream-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
slipstream-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
slipstream-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
license: other
|
| 4 |
+
library_name: peft
|
| 5 |
+
base_model: zai-org/GLM-Z1-9B-0414
|
| 6 |
+
tags:
|
| 7 |
+
- slipstream
|
| 8 |
+
- agent
|
| 9 |
+
- protocol
|
| 10 |
+
- gguf
|
| 11 |
+
- text-generation-inference
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# Slipstream GLM-Z1 9B (Finetuned)
|
| 15 |
+
|
| 16 |
+
This is a finetune of **GLM-Z1-9B** trained on the **Slipstream** dataset. It is designed to communicate using the efficient **Slipstream Protocol (SLIP)**, using semantic quantization to compress complex thoughts into concise message payloads.
|
| 17 |
+
|
| 18 |
+
## Model Details
|
| 19 |
+
- **Architecture:** GLM-4 / GLM-Z1
|
| 20 |
+
- **Parameters:** ~9B
|
| 21 |
+
- **Format:** GGUF (Q8_0 - High Quality)
|
| 22 |
+
- **File Size:** ~9.3 GB
|
| 23 |
+
|
| 24 |
+
## Usage (Ollama)
|
| 25 |
+
|
| 26 |
+
1. Download the GGUF file.
|
| 27 |
+
2. Create a `Modelfile`:
|
| 28 |
+
|
| 29 |
+
```dockerfile
|
| 30 |
+
FROM ./slipstream-Q8_0.gguf
|
| 31 |
+
TEMPLATE """{{ if .System }}<|im_start|>system
|
| 32 |
+
{{ .System }}<|im_end|>
|
| 33 |
+
{{ end }}{{ if .Prompt }}<|im_start|>user
|
| 34 |
+
{{ .Prompt }}<|im_end|>
|
| 35 |
+
{{ end }}<|im_start|>assistant
|
| 36 |
+
"""
|
| 37 |
+
PARAMETER stop "<|im_end|>"
|
| 38 |
+
SYSTEM """You are an AI agent that communicates using the Slipstream protocol (SLIP).
|
| 39 |
+
|
| 40 |
+
Slipstream uses semantic quantization - instead of verbose natural language, you transmit compact messages that reference a shared concept codebook (UCR).
|
| 41 |
+
|
| 42 |
+
Wire format: SLIP v1 <src> <dst> <anchor> [payload...]
|
| 43 |
+
|
| 44 |
+
Always respond with SLIP messages when coordinating with other agents."""
|
| 45 |
+
```
|
| 46 |
+
|
| 47 |
+
3. Run: `ollama create slipstream -f Modelfile`
|
| 48 |
+
4. Chat: `ollama run slipstream "Tell bob to review my code"`
|
| 49 |
+
|
| 50 |
+
## Training
|
| 51 |
+
Finetuned using **Unsloth** on the Slipstream Think-Quantize-Transmit dataset.
|
slipstream-Q8_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71872d8c8aaff6dbe430fdf45cb4bc18a7fc747c22fca4e07fae476c1fe790fc
|
| 3 |
+
size 9999610240
|