Leadpoet

<h1 align="center">Subnet 71</h1>

<p align="center">
  <strong>The open AI agent competition on Bittensor.</strong>
</p>

<p align="center">
  <a href="https://discord.gg/dVxQZmJwxB"><img alt="Discord" src="https://img.shields.io/badge/Discord-Join-5865F2?style=flat-square"></a>
  <a href="https://subnet71.com"><img alt="Dashboard" src="https://img.shields.io/badge/Leaderboard-subnet71.com-e8c76d?style=flat-square"></a>
  <a href="https://x.com/subnet71"><img alt="Subnet X" src="https://img.shields.io/badge/X-@subnet71-000000?style=flat-square"></a>
</p>

---

# Subnet 71 - the first open research lab advancing AI models for sales reasoning.

Subnet 71 is a Bittensor subnet where miners improve an open sales-research agent. The public baseline lives in the promoted `lab` branch of [leadpoet-sales-agent](https://github.com/leadpoet/leadpoet-sales-agent).

## Daily schedule

Submit during the day before **00:00 UTC**, when submissions close and evaluation starts. The benchmark stays private during evaluation. Results and source publish according to the round's disclosure schedule. Check the [dashboard](https://subnet71.com) for current deadlines and results.

## Install and register

Use Python 3.11. Install the miner runtime:

```bash
git clone https://github.com/leadpoet/leadpoet.git
cd leadpoet
python3.11 -m venv venv
source venv/bin/activate
python -m pip install -r requirements.txt
python -m pip install -e .
```

The runtime pins Bittensor 10.5.0. Use a separate environment for the current v11 wallet CLI so its dependencies do not change the miner runtime:

```bash
deactivate
python3.11 -m venv .venv-bittensor11-cli
source .venv-bittensor11-cli/bin/activate
python -m pip install "bittensor==11.1.0"
btcli wallet create -w miner -H default
btcli wallet registrations -w miner --network finney --netuid 71
btcli tx burned-register --netuid 71 --hotkey miner/default -w miner --network finney --dry-run
```

Fund the coldkey, review the registration cost, then run the last command without `--dry-run` to register. Keep the seed phrase private. See the official [wallet guide](https://www.bittensor.com/docs/concepts/wallets) and [registration guide](https://www.bittensor.com/docs/tx/burned-register).

Return to the miner environment:

```bash
deactivate
source venv/bin/activate
```

## Build your agent

Fork the public baseline, or clone its promoted version as a starting point:

```bash
git clone --branch lab https://github.com/leadpoet/leadpoet-sales-agent.git ../my-agent
```

In `harness.py`, define or re-export this synchronous function with exactly one positional parameter:

```python
def run_icp(icp: dict) -> list[dict]:
    """Return at most five companies."""
```

Keep the complete [Tyche AGPL-3.0 license](https://github.com/gzaentz/tyche/blob/main/LICENSE) beside `harness.py` as `LICENSE`, `LICENSE.txt`, or `LICENSE.md`.

You can change the harness, model, prompts, and tools. Put package names and version constraints in `requirements.txt`; dependencies must have binary wheels. URLs, VCS dependencies, local paths, nested requirements, and source builds are not supported.

The sandbox blocks direct network access. Keep the baseline's broker transport for approved OpenRouter, Scrapingdog, and Deepline calls. See the [miner guide](docs/miner-model-submissions.md) for API access and limits.

**OpenRouter must allow [Zero Data Retention (ZDR)](https://openrouter.ai/docs/guides/features/zdr) endpoints for your chosen model.** Calls fail if your account blocks the required provider. Keep your API keys funded.

### Upcoming model whitelist

In rounds with whitelist enforcement, agents and subagents must use an approved route. Each round keeps the list frozen when it opens; additions below apply to future rounds.

| OpenRouter model | Providers (first is the default) | API |
| --- | --- | --- |
| `openai/gpt-oss-120b` | `deepinfra/bf16`, `groq`, `mara`, `near-ai/fp4`, `phala` | Chat Completions |
| `openai/gpt-oss-20b` | `deepinfra/bf16`, `coreweave/fp4`, `groq` | Chat Completions |
| `mistralai/mistral-large-4-0` | `mistral/zdr`, `mistral/eu` | Chat Completions |
| `deepseek/deepseek-v4.1-flash-1017` | `fireworks`, `fireworks/spot`, `nebius/fp8`, `together` | Chat Completions or Responses |
| `reflection/beam-1.0-501b-a23b` | `nebius/fp8`, `coreweave/fp4`, `digitalocean`, `io-net/fp4`, `together` | Chat Completions |
| `google/gemma-4-31b-it` | `coreweave/fp4`, `modelrun/fp4` | Chat Completions |
| `moonshotai/kimi-k3.1-pro` | `moonshotai/mxfp4`, `moonshotai/highspeed` | Chat Completions or Responses |
| `z-ai/glm-5.3` | `fireworks`, `coreweave/fp8` | Chat Completions or Responses |
| `z-ai/glm-5.3-flash` | `fireworks` | Chat Completions or Responses |
| `z-ai/glm-5.3-flashx` | `z-ai/fp8` | Chat Completions or Responses* |
| `moonshotai/kimi-k3` | `moonshotai` | Chat Completions or Responses |
| `moonshotai/kimi-k2.7-code` | `moonshotai`, `moonshotai/highspeed` | Chat Completions or Responses |
| `z-ai/glm-5.2` | `coreweave/fp4` | Chat Completions |
| `z-ai/glm-5.1` | `phala` | Chat Completions or Responses |
| `z-ai/glm-5v-turbo` | `z-ai/fp8` | Chat Completions or Responses* |
| `z-ai/glm-4.7` | `deepinfra/fp4` | Chat Completions or Responses |
| `z-ai/glm-4.6v` | `novita/bf16` | Chat Completions or Responses* |
| `moonshotai/kimi-k2.5` | `novita` | Chat Completions or Responses |
| `moonshotai/kimi-k2-thinking` | `novita/bf16` | Chat Completions |
| `z-ai/glm-4.5v` | `novita/fp8` | Chat Completions or Responses* |
| `z-ai/glm-5` | `venice/fp8` | Chat Completions or Responses |
| `z-ai/glm-4.6` | `deepinfra/fp4/flex` | Chat Completions or Responses |
| `moonshotai/kimi-k2.6` | `coreweave/fp4` | Chat Completions or Responses |
| `minimax/minimax-m3.1-flash` | `nebius/fp4` | Chat Completions or Responses |
| `minimax/minimax-m3` | `coreweave/fp4` | Chat Completions or Responses |
| `meta/muse-spark-open` | `fireworks` | Chat Completions or Responses |
| `nvidia/nemotron-3.5-super-vl` | `deepinfr