> ## Documentation Index
> Fetch the complete documentation index at: https://learn.getodin.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Hoshi 1.0

> Our proprietary frontier model — agent-grade intelligence using a fraction of the credits

**Hoshi 1.0** is our proprietary frontier model, purpose-built for agentic workloads. It delivers frontier-tier performance on coding, structured output, and tool-use benchmarks at a **fraction of the credits** of leading alternatives.

## Why Hoshi 1.0?

Hoshi 1.0 is designed from the ground up for the kinds of tasks Odin AI Platform agents perform every day: reasoning over documents, calling tools, generating structured output, and following multi-step instructions. It is not a general-purpose chatbot — it is an **agent engine**.

<CardGroup cols={3}>
  <Card title="Credits, Not Dollars" icon="coins">
    Frontier-level quality at a fraction of the credits of GPT 5.5 or Sonnet 5.
  </Card>

  <Card title="OpenAI-Compatible" icon="code">
    Drop-in Chat Completions API — change three lines and keep your stack.
  </Card>

  <Card title="No Lock-In" icon="lock-open">
    Standard request/response shape. Switch back or route around us anytime.
  </Card>
</CardGroup>

## Pricing

Hoshi 1.0 uses a simple credit-based pricing model. No subscription is required for raw API access.

| Tier             | Credits                                         |
| ---------------- | ----------------------------------------------- |
| **Input**        | 150 / 1M tokens                                 |
| **Cached Input** | 10 / 1M tokens (\~15× cheaper than fresh input) |
| **Output**       | 300 / 1M tokens                                 |

<Note>
  Credits are the universal unit across Shogo. Agent plans (Free / Basic / Pro / Business) include a monthly credit allotment — see the [pricing page](https://www.shogo.ai/pricing) for details.
</Note>

## When to Use Hoshi vs. Frontier Models

Choosing the right model depends on the task. Use the guide below to decide when Hoshi 1.0 is the best fit and when you might reach for a frontier model like **Sonnet 5** or **GPT 5.5**.

### Use Hoshi 1.0 When…

| Scenario                             | Why Hoshi Wins                                                                    |
| ------------------------------------ | --------------------------------------------------------------------------------- |
| **Tool calling & function use**      | Purpose-tuned for reliable, low-latency tool invocations — the core of any agent. |
| **Structured output (JSON, tables)** | Consistently produces valid, schema-conformant output without retries.            |
| **Multi-step reasoning**             | Excels at chaining actions, parsing responses, and deciding next steps.           |
| **High-volume batch tasks**          | At 150 credits / 1M input tokens, running thousands of agent turns is economical. |
| **Knowledge Base retrieval**         | Optimized for RAG-style workflows — grounding answers in your documents.          |
| **Cost-sensitive production**        | Same quality tier as frontier models at a fraction of the credits.                |
| **Code generation & refactoring**    | Sweeps coding benchmarks; generates, edits, and reviews code reliably.            |

### Use a Frontier Model When…

| Scenario                                 | Why a Frontier Model May Be Better                                                                       |
| ---------------------------------------- | -------------------------------------------------------------------------------------------------------- |
| **Ultra-long context (> 200K tokens)**   | Models like Claude Sonnet 5 or GPT 5.5 support very large context windows for massive document analysis. |
| **Cutting-edge multimodal reasoning**    | When you need best-in-class vision, audio, or cross-modal understanding.                                 |
| **Highly creative / open-ended writing** | Frontier models may produce more varied, stylistically rich prose for marketing or storytelling.         |
| **Regulatory or compliance constraints** | Some organizations require a specific provider's model for data governance reasons.                      |
| **Research-grade tasks**                 | When you need the absolute latest capability on a novel, unstructured problem.                           |

<Warning>
  For most Odin AI Platform agent workloads — tool calling, structured output, RAG, and multi-step automation — **Hoshi 1.0 is the recommended default**. Use frontier models only when you have a specific reason to do so.
</Warning>

## Quick Comparison

| Dimension             | Hoshi 1.0        | Sonnet 5     | GPT 5.5      |
| --------------------- | ---------------- | ------------ | ------------ |
| **Input cost**        | 150 credits / 1M | \$2.00 / 1M  | \$5.00 / 1M  |
| **Output cost**       | 300 credits / 1M | \$10.00 / 1M | \$30.00 / 1M |
| **Tool calling**      | ★★★★★            | ★★★★★        | ★★★★★        |
| **Structured output** | ★★★★★            | ★★★★☆        | ★★★★☆        |
| **Coding**            | ★★★★★            | ★★★★★        | ★★★★★        |
| **Creative writing**  | ★★★★☆            | ★★★★★        | ★★★★★        |
| **Multimodal**        | ★★★☆☆            | ★★★★★        | ★★★★★        |
| **Context window**    | 128K             | 200K         | 128K         |
| **OpenAI-compatible** | ✅                | ❌            | ✅            |

## Getting Started

Hoshi 1.0 is available in Odin AI Platform as a predefined model. No API key configuration is required.

### Select Hoshi in Your Agent

<Steps>
  <Step title="Navigate to Agents">
    Go to **Agents** in the sidebar.
  </Step>

  <Step title="Open or create an agent">
    Select an existing agent or create a new one.
  </Step>

  <Step title="Go to the General tab">
    Open the **General** tab in the agent configuration.
  </Step>

  <Step title="Select Hoshi 1.0">
    Find the **Model** drop-down selector and choose **Hoshi 1.0** from the list.
  </Step>
</Steps>

### Use via API

Hoshi 1.0 is fully OpenAI-compatible. Point your existing client at the Shogo API:

```bash theme={null}
curl https://api.shogo.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hoshi-1.0",
    "messages": [
      {"role": "user", "content": "Summarize this document into bullet points."}
    ]
  }'
```

Or use the OpenAI SDK:

```python theme={null}
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.shogo.ai/v1"
)

response = client.chat.completions.create(
    model="hoshi-1.0",
    messages=[{"role": "user", "content": "Summarize this document into bullet points."}]
)
print(response.choices[0].message.content)
```

## Routing: Use Hoshi by Default, Frontier by Exception

Odin AI Platform's model routing lets you set **Hoshi 1.0 as your default model** across all agents and selectively override to a frontier model only when a specific task demands it. This gives you the best of both worlds:

1. **Default to Hoshi** — all agents run on Hoshi 1.0 for cost-efficient, high-quality performance.
2. **Override per agent** — switch individual agents to Sonnet 5 or GPT 5.5 when the task requires it (e.g., multimodal analysis, ultra-long context).
3. **Monitor and adjust** — track performance and cost per agent, then optimize your model mix over time.

## FAQ

<AccordionGroup>
  <Accordion title="Is Hoshi 1.0 as good as Sonnet 5 or GPT 5.5?">
    For agentic tasks — tool calling, structured output, coding, and multi-step reasoning — Hoshi 1.0 performs at the same frontier tier. Frontier models may have an edge on highly creative writing, multimodal reasoning, and extremely long contexts.
  </Accordion>

  <Accordion title="Can I use Hoshi 1.0 with my own API key?">
    Yes. Hoshi 1.0 is available both as a predefined model in Odin AI Platform (no key needed) and via the raw API with your own Shogo API key.
  </Accordion>

  <Accordion title="What is cached input pricing?">
    When you send the same context repeatedly (e.g., the same system prompt or knowledge base retrieval), cached tokens are billed at 10 credits / 1M — roughly 15× cheaper than fresh input tokens.
  </Accordion>

  <Accordion title="How does Hoshi handle data privacy?">
    Hoshi 1.0 runs on Shogo infrastructure. Your data is not used to train the model. For full details, see the [Shogo privacy policy](https://www.shogo.ai/privacy).
  </Accordion>

  <Accordion title="Will Hoshi 1.0 be deprecated soon?">
    No. Hoshi 1.0 is our flagship model and will be actively maintained and improved. Future versions will be announced with clear migration timelines.
  </Accordion>
</AccordionGroup>

## Learn More

* [Hoshi 1.0 Product Page](https://www.shogo.ai/hoshi/) — full technical details and benchmarks
* [Model Pricing](/agents/models) — compare all available models and their costs
* [Agent Best Practices](/agents/best-practices) — tips for building effective agents