Text Generation
Transformers
Safetensors
English
llama
vllm
insurance
reinsurance
sovereign-ai
llama-3
conversational
text-generation-inference
Instructions to use Reinsured-AI/Reinsure-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Reinsured-AI/Reinsure-8B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Reinsured-AI/Reinsure-8B") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Reinsured-AI/Reinsure-8B") model = AutoModelForCausalLM.from_pretrained("Reinsured-AI/Reinsure-8B", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Reinsured-AI/Reinsure-8B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Reinsured-AI/Reinsure-8B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Reinsured-AI/Reinsure-8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Reinsured-AI/Reinsure-8B
- SGLang
How to use Reinsured-AI/Reinsure-8B with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Reinsured-AI/Reinsure-8B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Reinsured-AI/Reinsure-8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Reinsured-AI/Reinsure-8B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Reinsured-AI/Reinsure-8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use Reinsured-AI/Reinsure-8B with Docker Model Runner:
docker model run hf.co/Reinsured-AI/Reinsure-8B
Sovereign AI B2B Marketing Copy Update
Browse files
README.md
CHANGED
|
@@ -5,35 +5,86 @@ tags:
|
|
| 5 |
- vllm
|
| 6 |
- insurance
|
| 7 |
- reinsurance
|
|
|
|
|
|
|
| 8 |
library_name: transformers
|
| 9 |
pipeline_tag: text-generation
|
| 10 |
---
|
| 11 |
|
| 12 |
-
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
-
|
| 17 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
-
|
| 20 |
|
| 21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
-
### Example vLLM Deployment (Modal)
|
| 24 |
```python
|
| 25 |
import modal
|
| 26 |
|
| 27 |
vllm_image = modal.Image.debian_slim().pip_install("vllm")
|
| 28 |
-
app = modal.App("reinsure-8b")
|
| 29 |
|
| 30 |
-
@app.function(
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
@modal.web_server(8000)
|
| 32 |
def serve():
|
| 33 |
import sys, subprocess
|
| 34 |
subprocess.Popen([
|
| 35 |
sys.executable, "-m", "vllm.entrypoints.openai.api_server",
|
| 36 |
"--model", "Reinsured-AI/Reinsure-8B",
|
| 37 |
-
"--port", "8000"
|
|
|
|
| 38 |
])
|
| 39 |
```
|
|
|
|
|
|
|
|
|
| 5 |
- vllm
|
| 6 |
- insurance
|
| 7 |
- reinsurance
|
| 8 |
+
- sovereign-ai
|
| 9 |
+
- llama-3
|
| 10 |
library_name: transformers
|
| 11 |
pipeline_tag: text-generation
|
| 12 |
---
|
| 13 |
|
| 14 |
+
<div align="center">
|
| 15 |
+
<h1>Reinsure-8B</h1>
|
| 16 |
+
<h3>The world's first insurance-native language model</h3>
|
| 17 |
+
</div>
|
| 18 |
|
| 19 |
+
<p align="center">
|
| 20 |
+
<a href="https://www.reinsured.ai/reinsure-8b">View Website</a> |
|
| 21 |
+
<a href="https://www.reinsured.ai/contact">Request API Access</a> |
|
| 22 |
+
<a href="https://www.reinsured.ai/platform">Platform Architecture</a>
|
| 23 |
+
</p>
|
| 24 |
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
## The world's first insurance-native language model
|
| 28 |
+
|
| 29 |
+
**Reinsure-8B** is a small language model built exclusively for the reinsurance and insurance industry. Fine-tuned from Llama 3.1 on reinsurance workflows, treaty structures, bordereaux formats, and London Market language.
|
| 30 |
+
|
| 31 |
+
Deploy it sovereign inside your own infrastructure, or call it as an inference API. Either way, you get a model that speaks insurance β without prompt engineering, without hallucinated policy terms.
|
| 32 |
+
|
| 33 |
+
### What is Organisational Sovereign AI?
|
| 34 |
+
Sovereign AI means the model runs inside your control. Your weights, your infrastructure, your data. No dependency on a third-party API that can change pricing, deprecate versions, or inspect your query traffic.
|
| 35 |
+
|
| 36 |
+
For regulated insurance businesses β Lloyd's syndicates, global reinsurers, captives, MGAs β sovereignty is not a preference. It is a compliance requirement. Sensitive submissions, treaty terms, and client data cannot flow to shared cloud endpoints.
|
| 37 |
+
|
| 38 |
+
Reinsure-8B is purpose-sized for sovereign deployment. At 8 billion parameters, it runs efficiently on enterprise GPU hardware β a single A100, L4, or equivalent β without the infrastructure overhead of 70B+ models. This is the practical path to production AI in a regulated industry.
|
| 39 |
+
|
| 40 |
+
### Two ways to run Reinsure-8B
|
| 41 |
+
1. **Sovereign Deployment:** Deploy the full model weights inside your own cloud or on-prem environment. Fine-tune it on your proprietary data. Complete data sovereignty and maximum performance.
|
| 42 |
+
2. **Hosted API:** Call Reinsure-8B as a hosted API via Reinsured.AI. No infrastructure required. Pay per inference token. Ideal for teams validating use cases or building lightweight integrations.
|
| 43 |
+
|
| 44 |
+
### Built for the language of reinsurance
|
| 45 |
+
Generic large language models are trained on the open internet β predominantly consumer content, code, and general text. Insurance knowledge is sparse, often incorrect, and never updated with current market practice.
|
| 46 |
|
| 47 |
+
Reinsure-8B was fine-tuned on a curated corpus of reinsurance-specific content β treaty wordings, bordereaux templates, Lloyd's market standards, catastrophe model outputs, underwriting guidelines, and claims documentation.
|
| 48 |
|
| 49 |
+
The result is a model that interprets reinsurance language correctly by default, without requiring you to explain what a "binder", "burning cost", or "cedant" means in every prompt.
|
| 50 |
+
|
| 51 |
+
### Domain-specific beats general-purpose
|
| 52 |
+
Applying a general-purpose model to insurance creates compounding problems β hallucination, data risk, poor economics. Reinsure-8B is purpose-built to eliminate each of them.
|
| 53 |
+
|
| 54 |
+
### The intelligence engine behind the stack
|
| 55 |
+
Reinsure-8B is the reasoning core that powers Reinsured.AI's Context Cloud and AI Agents. When an agent interprets a treaty clause, extracts a bordereaux field, or classifies a submission, it is calling on Reinsure-8B β a model that already understands the domain.
|
| 56 |
+
|
| 57 |
+
Organisations that deploy Reinsure-8B sovereign get the additional option to fine-tune it on their own internal data, creating a model layer unique to their underwriting philosophy and market positioning β one that becomes a proprietary asset over time.
|
| 58 |
+
|
| 59 |
+
---
|
| 60 |
+
|
| 61 |
+
## Technical Specifications & Deployment
|
| 62 |
+
|
| 63 |
+
This model has been exported to standard **16-bit Safetensors** format. It is fully compatible with industry-standard cloud inference engines, including **vLLM** and HuggingFace Text Generation Inference (TGI).
|
| 64 |
+
|
| 65 |
+
### Cloud Native Example (vLLM / Modal)
|
| 66 |
+
For organizations deploying Sovereign AI at scale with scale-to-zero economics, Reinsure-8B is optimized for `vLLM`. Here is a reference architecture deploying to [Modal Serverless](https://modal.com):
|
| 67 |
|
|
|
|
| 68 |
```python
|
| 69 |
import modal
|
| 70 |
|
| 71 |
vllm_image = modal.Image.debian_slim().pip_install("vllm")
|
| 72 |
+
app = modal.App("reinsure-8b-sovereign")
|
| 73 |
|
| 74 |
+
@app.function(
|
| 75 |
+
image=vllm_image,
|
| 76 |
+
gpu="L4", # Extremely cost-effective for 8B models
|
| 77 |
+
container_idle_timeout=300 # Scale to zero when inactive
|
| 78 |
+
)
|
| 79 |
@modal.web_server(8000)
|
| 80 |
def serve():
|
| 81 |
import sys, subprocess
|
| 82 |
subprocess.Popen([
|
| 83 |
sys.executable, "-m", "vllm.entrypoints.openai.api_server",
|
| 84 |
"--model", "Reinsured-AI/Reinsure-8B",
|
| 85 |
+
"--port", "8000",
|
| 86 |
+
"--max-model-len", "8192"
|
| 87 |
])
|
| 88 |
```
|
| 89 |
+
|
| 90 |
+
[Book a Demo](https://www.reinsured.ai/demo) | [Contact the Team](https://www.reinsured.ai/contact)
|