Text Generation
PEFT
Safetensors
English
deployxa
devops
sre
cloud
deployment-intelligence
root-cause-analysis
self-healing
container-orchestration
kubernetes
docker
conversational
Eval Results (legacy)
Instructions to use deployxa/deployxa-lm-1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use deployxa/deployxa-lm-1.0 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-Coder-1.5B-Instruct") model = PeftModel.from_pretrained(base_model, "deployxa/deployxa-lm-1.0") - Notebooks
- Google Colab
- Kaggle
docs: add Ollama badge, quickstart, and link to deployxa/cloud-sre-bench
Browse files
README.md
CHANGED
|
@@ -54,12 +54,13 @@ model-index:
|
|
| 54 |
**Build. Deploy. Scale.** • *Push code. Get a live URL.*
|
| 55 |
|
| 56 |
[](https://opensource.org/licenses/Apache-2.0)
|
|
|
|
| 57 |
[](https://pypi.org/project/dxa-lm/)
|
| 58 |
[](https://chat.deployxa.com)
|
| 59 |
[](https://deployxa.com)
|
| 60 |
[](https://deployxa.com)
|
| 61 |
|
| 62 |
-
[Official Platform](https://deployxa.com) • [Live Web Console](https://chat.deployxa.com) • [Documentation](https://docs.deployxa.com) • [Brand Kit](https://deployxa.com/brand) • [Benchmark Suite](https://github.com/deployxa/cloud-sre-bench)
|
| 63 |
|
| 64 |
</div>
|
| 65 |
|
|
@@ -187,7 +188,15 @@ All evaluations are conducted against an open dataset of 384 real deployment fai
|
|
| 187 |
|
| 188 |
## 🚀 Quickstart
|
| 189 |
|
| 190 |
-
### 1. Using
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 191 |
|
| 192 |
```bash
|
| 193 |
# Install the CLI from PyPI
|
|
@@ -206,7 +215,7 @@ dxa-lm chat
|
|
| 206 |
dxa-lm health
|
| 207 |
```
|
| 208 |
|
| 209 |
-
###
|
| 210 |
|
| 211 |
```python
|
| 212 |
import torch
|
|
@@ -283,11 +292,16 @@ Deployxa LM 1.0 enforces strict safety and reliability invariants:
|
|
| 283 |
|
| 284 |
## 🧪 Open Benchmark Suite
|
| 285 |
|
| 286 |
-
To ensure community reproducibility, the complete evaluation harness
|
| 287 |
|
| 288 |
-
Run the
|
| 289 |
```bash
|
| 290 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 291 |
```
|
| 292 |
|
| 293 |
---
|
|
|
|
| 54 |
**Build. Deploy. Scale.** • *Push code. Get a live URL.*
|
| 55 |
|
| 56 |
[](https://opensource.org/licenses/Apache-2.0)
|
| 57 |
+
[](https://ollama.com/deployxa/deployxa-lm)
|
| 58 |
[](https://pypi.org/project/dxa-lm/)
|
| 59 |
[](https://chat.deployxa.com)
|
| 60 |
[](https://deployxa.com)
|
| 61 |
[](https://deployxa.com)
|
| 62 |
|
| 63 |
+
[Official Platform](https://deployxa.com) • [Live Web Console](https://chat.deployxa.com) • [Ollama Model](https://ollama.com/deployxa/deployxa-lm) • [Documentation](https://docs.deployxa.com) • [Brand Kit](https://deployxa.com/brand) • [Benchmark Suite](https://github.com/deployxa/cloud-sre-bench)
|
| 64 |
|
| 65 |
</div>
|
| 66 |
|
|
|
|
| 188 |
|
| 189 |
## 🚀 Quickstart
|
| 190 |
|
| 191 |
+
### 1. Using Ollama (Instant 1-Command Local Run)
|
| 192 |
+
|
| 193 |
+
Run Deployxa LM 1.0 locally in Ollama:
|
| 194 |
+
|
| 195 |
+
```bash
|
| 196 |
+
ollama run deployxa/deployxa-lm
|
| 197 |
+
```
|
| 198 |
+
|
| 199 |
+
### 2. Using the Official Deployxa CLI (`dxa-lm`)
|
| 200 |
|
| 201 |
```bash
|
| 202 |
# Install the CLI from PyPI
|
|
|
|
| 215 |
dxa-lm health
|
| 216 |
```
|
| 217 |
|
| 218 |
+
### 3. Python SDK / In-Process Execution
|
| 219 |
|
| 220 |
```python
|
| 221 |
import torch
|
|
|
|
| 292 |
|
| 293 |
## 🧪 Open Benchmark Suite
|
| 294 |
|
| 295 |
+
To ensure community reproducibility, the complete evaluation harness, 26 real-world failure log cases, and 14 grounded platform queries are available in [`deployxa/cloud-sre-bench`](https://github.com/deployxa/cloud-sre-bench).
|
| 296 |
|
| 297 |
+
Run the benchmark evaluation suite:
|
| 298 |
```bash
|
| 299 |
+
git clone https://github.com/deployxa/cloud-sre-bench.git
|
| 300 |
+
cd cloud-sre-bench
|
| 301 |
+
|
| 302 |
+
# Run against Ollama or live API
|
| 303 |
+
python3 run_benchmark.py --target api
|
| 304 |
+
# or: python3 run_benchmark.py --target ollama --model deployxa/deployxa-lm
|
| 305 |
```
|
| 306 |
|
| 307 |
---
|