Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
datasets:
|
| 6 |
+
- continuedev/instinct-data
|
| 7 |
+
base_model:
|
| 8 |
+
- Qwen/Qwen2.5-Coder-7B
|
| 9 |
+
pipeline_tag: text-generation
|
| 10 |
+
library_name: transformers
|
| 11 |
+
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
[](https://hf.co/QuantFactory)
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
# QuantFactory/instinct-GGUF
|
| 18 |
+
This is quantized version of [continuedev/instinct](https://huggingface.co/continuedev/instinct) created using llama.cpp
|
| 19 |
+
|
| 20 |
+
# Original Model Card
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/686c5c546abedce0f7ac048a/B7PeaDQCDnlgT3Tmf7fsb.png" width=250>
|
| 24 |
+
|
| 25 |
+
# Instinct, the State-of-the-Art Open Next Edit Model
|
| 26 |
+
|
| 27 |
+
This repo contains the model weights for [Continue](https://continue.dev)'s state-of-the-art open Next Edit model, **Instinct**. Robustly fine-tuned from Qwen2.5-Coder-7B on our [dataset of real-world code edits](https://huggingface.co/datasets/continuedev/instinct-data), Instinct intelligently predicts your next move to keep you in flow.
|
| 28 |
+
|
| 29 |
+
## Serving the model
|
| 30 |
+
|
| 31 |
+
**Ollama**: We've released a [Q4_K_M GGUF quantization of Instinct](https://huggingface.co/continuedev/instinct-GGUF) for efficient local inference. Try it with [Continue's Ollama integration](https://docs.continue.dev/guides/ollama-guide), or just run `ollama run nate/instinct`.
|
| 32 |
+
|
| 33 |
+
You can also serve the model using either of the below options, then [connect it with Continue](https://docs.continue.dev/guides/how-to-self-host-a-model).
|
| 34 |
+
|
| 35 |
+
**SGLang**: `python3 -m sglang.launch_server --model-path continuedev/instinct --load-format safetensors`
|
| 36 |
+
<br>**vLLM**: `vllm serve continuedev/instinct --served-model-name instinct --load-format safetensors`
|
| 37 |
+
|
| 38 |
+
## Learn more
|
| 39 |
+
|
| 40 |
+
For more information on the work behind Instinct, please refer to our [blog](https://blog.continue.dev/instinct/).
|