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

# Pricing Overview

> Transparent, usage-based pricing aligned with your success

Tensormesh uses a transparent, usage-based pricing model designed to align with your success. You only pay for what you use, and our pricing reflects the value you receive.

***

## Serverless Token Pricing

Serverless inference is billed on a **per-token** basis. Each model has three token price tiers:

| Token Type       | What It Is                                                    | Cost           |
| ---------------- | ------------------------------------------------------------- | -------------- |
| **Input Token**  | Tokens in your prompt (system message, user message, context) | Per-model rate |
| **Output Token** | Tokens generated in the model's response                      | Per-model rate |
| **Cached Token** | Input tokens served from Tensormesh's KV cache                | **\$0.00**     |

<Note>
  **Cached tokens are free.** When Tensormesh serves a token from its KV cache instead of recomputing it, you are not charged. The more your requests share common prefixes — system messages, conversation history, shared context — the higher your cache hit rate and the lower your effective cost.
</Note>

Current per-model rates are displayed on the **Deploy → Serverless** page. Each model card shows the exact input, output, and cached token price.

***

## How \$0 Cached Tokens Work

Every time you send a request, Tensormesh checks whether the input tokens have already been computed and stored in the KV cache. If they have, those tokens are served instantly — at zero cost to you.

<CardGroup cols={2}>
  <Card title="First Request" icon="microchip">
    Input tokens are computed and stored in the KV cache. You pay the standard input token rate.
  </Card>

  <Card title="Subsequent Requests" icon="bolt">
    Matching input tokens are served from cache. **You pay \$0** for those tokens.
  </Card>
</CardGroup>

### What Gets Cached

* System messages and instructions
* Shared conversation prefixes
* Repeated context windows (e.g. long documents passed in every request)
* Common prompt templates

### Maximizing Your Cache Hit Rate

<AccordionGroup>
  <Accordion title="Use consistent system messages" icon="bullseye">
    Keep your system prompt identical across requests. Even a single character change creates a cache miss. Use a fixed template and only vary the user message.
  </Accordion>

  <Accordion title="Put stable content first" icon="layer-group">
    Structure prompts so that the content that changes least (system prompt, static context) comes before content that varies (the user's latest message). The cache matches from the start of the prompt.
  </Accordion>

  <Accordion title="Standardize prompt templates" icon="wand-magic-sparkles">
    Implement shared prompt templates across your application. Consistent prefixes across different users or sessions all benefit from the same cached tokens.
  </Accordion>

  <Accordion title="Monitor your hit rate" icon="chart-line">
    Check your cache hit rate under **Operations → Serverless Usage**. If it's low, look for variability in the early parts of your prompts — that's where cache misses happen.
  </Accordion>
</AccordionGroup>

***

## External Storage

By default, the KV cache is in-memory and scoped to a single request window. **External Storage** extends the cache to a persistent bucket that survives across requests, sessions, and even across different users in your application.

<Info>
  External Storage is available as a paid subscription tier. Navigate to **Operations → Storage** to view plans and subscribe.
</Info>

### What External Storage Does

Without external storage, every new session starts cold — tokens that were cached in a previous session need to be recomputed. With a storage bucket, that context is persisted.

<CardGroup cols={2}>
  <Card title="Persistent KV Cache" icon="database">
    Cache entries survive beyond a single request window. Long system prompts and shared context are stored across sessions.
  </Card>

  <Card title="Cross-Session Savings" icon="repeat">
    Users returning to an existing conversation, or requests sharing a common prefix, all benefit from cached tokens — even hours later.
  </Card>

  <Card title="Faster Cold Starts" icon="bolt">
    New sessions that share context with previous ones skip recomputation entirely, reducing time-to-first-token.
  </Card>

  <Card title="Works Automatically" icon="check">
    Once your bucket is active, caching is handled transparently. No changes to your API calls required.
  </Card>
</CardGroup>

### Storage Tiers

Storage plans are tiered by bucket size. A **free tier (0 GB)** is available by default — you only pay when you subscribe to a plan.

| Plan       | Best For                                                          |
| ---------- | ----------------------------------------------------------------- |
| **Bronze** | Getting started — low to moderate request volume                  |
| **Silver** | Agentic developers — more headroom for parallel workloads         |
| **Gold**   | Production-scale inference — high volume and large system prompts |

Subscribe or change plans anytime under **Operations → Storage**. Billing adjusts immediately — no data is lost when upgrading.

***

## Billing Details

<CardGroup cols={2}>
  <Card title="Billing Frequency" icon="calendar">
    Monthly billing cycles with detailed usage reports
  </Card>

  <Card title="Cost Tracking" icon="chart-mixed">
    Real-time cost visibility under Operations → Serverless Usage
  </Card>

  <Card title="Usage Analytics" icon="magnifying-glass-chart">
    Per-model breakdown of input, output, and cached token counts
  </Card>

  <Card title="Storage Subscription" icon="database">
    Flat monthly rate per storage tier, billed separately from token usage
  </Card>
</CardGroup>

***

## Frequently Asked Questions

<AccordionGroup>
  <Accordion title="Are cached tokens really free?">
    Yes. When a token is served from the KV cache, the cached token price is \$0.00. You are only charged the standard input token rate for tokens that need to be freshly computed.
  </Accordion>

  <Accordion title="How do I know how many tokens are being cached?">
    Navigate to **Operations → Serverless Usage** to see your input, output, and cached token counts per model and per period.
  </Accordion>

  <Accordion title="Do I need External Storage to get $0 cached tokens?">
    No. In-memory caching is active by default and cached tokens are always \$0.00. External Storage extends caching across sessions and increases your cache hit rate, but it is not required to benefit from \$0 cached tokens.
  </Accordion>

  <Accordion title="What happens if my cache hit rate is 0%?">
    You pay only the standard input and output token rates. There are no additional charges. A 0% cache hit rate just means you're not saving anything yet — structuring prompts with consistent prefixes is the fastest way to improve it.
  </Accordion>

  <Accordion title="Are there any hidden fees?">
    No. You pay per token for serverless inference plus an optional flat monthly rate for External Storage. No setup fees, no minimums, no per-request overhead.
  </Accordion>

  <Accordion title="Can I predict my costs?">
    Check the per-model rates on the **Deploy → Serverless** page, estimate your input and output token counts, and factor in your expected cache hit rate. Your Serverless Usage history is a good baseline for projecting future spend.
  </Accordion>
</AccordionGroup>