You can deploy this model and start earning money today!

bobroller125/
Supreme_v2

Name: Supreme_v2
Availability: InStock

Views

No views yet

Supreme_V2 — Subnet 5 Model Provider

Model: Supreme_V2
Architecture: Mistral 7B
Usage: Bittensor Subnet 5 (HONE / Manifold Labs)
Status: Production-ready
License: Apache 2.0

🧠 Overview

Supreme_V2 is a customized Mistral 7B model optimized specifically for Bittensor Subnet 5.
It has been configured for:

stable and consistent responses
fast inference on consumer GPUs
deterministic behavior required by Subnet 5 validators
full compatibility with the HONE Sandbox

The model is provided in the safetensors format, split into 3 shards to ensure maximum compatibility during streaming and loading.

📁 Included Files

Model Weights

model-00001-of-00003.safetensors
model-00002-of-00003.safetensors
model-00003-of-00003.safetensors
model.safetensors.index.json

Configuration

config.json
generation_config.json
special_tokens_map.json
tokenizer.model
tokenizer_config.json

Sandbox (Required for Subnet 5)

sandbox/sandbox_runner.py
sandbox/requirements.txt
sandbox/__init__.py

These files allow Subnet 5 validators to load the model deterministically and evaluate it during inference tasks.

🔧 Sandbox Entrypoint

Subnet 5 validators will load the model using the following interface:

python

from sandbox import SandboxRunner

runner = SandboxRunner()
response = runner.infer("Hello!")
print(response)

bobroller125/Supreme_v2