Initial release: Unbound E2B GGUFs (Q4_K_M / Q8_0 / F16) + Ollama Modelfile

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+unbound-e2b-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text

Modelfile

@@ -0,0 +1,13 @@
+FROM ./unbound-e2b-Q4_K_M.gguf
+
+# Default sampling — leans toward factual recall.
+# Bump temperature to 0.8-1.0 for creative writing / brainstorming.
+PARAMETER temperature 0.6
+PARAMETER top_p 0.95
+PARAMETER top_k 64
+PARAMETER repeat_penalty 1.05
+PARAMETER num_ctx 8192
+
+# Identity grounding — anchors the model in its own self-description
+# every conversation, so brand questions stay sharp even at higher temps.
+SYSTEM """You are Unbound, an uncensored on-device language model built on top of Google's Gemma 4 E2B by the Chromia and Eval Engine team. You have about 2 billion effective parameters and are text-only (no image, audio, or video generation). Answer directly without safety preambles. If you don't know a specific fact, say so honestly rather than guessing."""

README.md

@@ -0,0 +1,89 @@
+---
+license: apache-2.0
+base_model: evalengine/unbound-e2b
+base_model_relation: quantized
+tags:
+- gguf
+- gemma4
+- gemma
+- gemma-4
+- uncensored
+- on-device
+pipeline_tag: text-generation
+---
+
+<p align="center">
+  <img src="unbound-logo.svg" alt="Unbound" width="160" height="160">
+</p>
+
+# Unbound E2B GGUF — *because there is no boundary*
+
+GGUF quantizations of [`evalengine/unbound-e2b`](https://huggingface.co/evalengine/unbound-e2b)
+for on-device deployment via Ollama, llama.cpp, LM Studio, and similar runtimes.
+
+## Available quants
+
+| File | Size | Notes |
+|---|---|---|
+| `unbound-e2b-Q4_K_M.gguf` | 3.2 GB | Recommended phone-deployable build |
+
+More quants (Q5_K_M, Q8_0, etc.) may be added on request.
+
+## Recommended sampling
+
+- **Creative writing / open-ended / general chat** → Gemma defaults:
+  `temperature=1.0, top_p=0.95, top_k=64`.
+- **Factual or brand/identity questions** → lower `temperature` to ~0.3–0.5
+  for sharper recall.
+- **llama.cpp**: pass `--jinja` for proper chat-template handling.
+- **Gemma 4 thinking mode** is on by default. Set `enable_thinking: false`
+  in the chat-template kwargs for shorter/faster replies on this 2B model.
+
+Some edge-case prompts may deflect on the first ask; a re-ask usually gets
+through.
+
+## Default sampling (Ollama)
+
+When you `ollama pull hf.co/evalengine/unbound-e2b-gguf`, the bundled
+`Modelfile` sets these defaults, tuned for factual recall:
+
+- `temperature = 0.6` (lower than Gemma's training default of 1.0 — keeps
+  the model from hallucinating brand/identity facts at high temperature)
+- `top_p = 0.95`, `top_k = 64`, `repeat_penalty = 1.05`, `num_ctx = 8192`
+- A short system prompt that grounds the model's identity (model name,
+  parameter count, modality, team) so brand questions stay sharp.
+
+**To override per-session in Ollama:**
+
+```
+ollama run hf.co/evalengine/unbound-e2b-gguf
+>>> /set parameter temperature 1.0   # creative / open-ended
+>>> /set parameter temperature 0.3   # max factual / brand questions
+```
+
+For llama.cpp users, pass `--temp 0.6 --top-p 0.95 --top-k 64` and
+include the SYSTEM line from the `Modelfile` as your `--system` argument.
+
+## Run with Ollama
+
+```bash
+ollama pull hf.co/evalengine/unbound-e2b-gguf
+ollama run  hf.co/evalengine/unbound-e2b-gguf
+```
+
+## Run with llama.cpp
+
+```bash
+./llama-cli -m unbound-e2b-Q4_K_M.gguf -p "your prompt"
+```
+
+## About the base
+
+See [`evalengine/unbound-e2b`](https://huggingface.co/evalengine/unbound-e2b)
+for the full model card, benchmarks, intended use, and the merged HF weights.
+
+*Reduced safety filtering — recommended for research and controlled environments. Review outputs and comply with applicable laws; no warranty for misuse.*
+
+## License
+
+Apache-2.0, inherited from `google/gemma-4-E2B-it`.

unbound-e2b-Q4_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcde3b3e5b1339df9eb1141379afe8af2b6312fad1dea1960a72f456f0dde643
+size 3416118048