Skip to content
HIPS
  • Earn with your GPU
  • Pricing
  • Docs
Sign inJoin early access
⌘K

Get started

  1. 01Developer hub
  2. 02Quickstart
  3. 03Migrate to HIPS

Guides

  1. 04HIPS SDK
  2. 05Import and run your model
  3. 06Model and task guides
    1. ATask guides
    2. BLanguage (chat)
    3. CClassification
    4. DImage
    5. EVideo
    6. FAudio
    7. GDocument processing
  4. 07Batch jobs
  5. 08Prompt caching
  6. 09Code with HIPS
  7. 10Coding-tool setup
    1. AHIPS Code
    2. BClaude Code
    3. CCodex
    4. DOpenCode
    5. EKilo
    6. FPi
    7. GQwen Code
    8. HCrush
    9. ICline
    10. JPrime Agent
    11. KAider
    12. LGoose
    13. MHermes Agent
    14. NGrok Build
    15. ODeepSeek Harness
    16. POther coding tools and editor integrations
  8. 11Command line
    1. AInstall and update
    2. BSign in
    3. CStatus and models
    4. DCoding agents
  9. 12Console
    1. AAccount
    2. BWorkspaces
    3. CAPI keys
    4. DSettings
    5. EUsage
    6. FModels
    7. GPlayground
    8. HDeployments
  10. 13Billing
  11. 14Provider setup
    1. AWhen it works
    2. BCommands
    3. CSettings
    4. DProvider Console
    5. EEarnings
    6. FCrash reports
    7. GData centers
  12. 15Semantic routing
  13. 16Guardrails
  14. 17Knowledge

Reference

  1. 18API reference
    1. AAuthentication
    2. BChat completions
    3. CResponses
    4. DMessages
    5. ECount tokens
    6. FModels
    7. GPrices API
    8. HTasks
    9. IEmbeddings and rerank
    10. JImages
    11. KJobs
    12. LUploads
    13. MFiles
    14. NBatches
    15. OWebhooks
    16. PVector stores
    17. QDeployments
    18. RRequirements and headers
    19. SErrors
    20. TUsage
    21. UVersioning
  2. 19Operating guidance
    1. ALimits
    2. BErrors
    3. CRetries and billing outcomes
    4. DIdempotency
    5. EStreaming
    6. FData handling
    7. GCompatibility matrix
  3. 20Changelog
Sheet indexOperating guidance

Get started

  1. 01Developer hub
  2. 02Quickstart
  3. 03Migrate to HIPS

Guides

  1. 04HIPS SDK
  2. 05Import and run your model
  3. 06Model and task guides
    1. ATask guides
    2. BLanguage (chat)
    3. CClassification
    4. DImage
    5. EVideo
    6. FAudio
    7. GDocument processing
  4. 07Batch jobs
  5. 08Prompt caching
  6. 09Code with HIPS
  7. 10Coding-tool setup
    1. AHIPS Code
    2. BClaude Code
    3. CCodex
    4. DOpenCode
    5. EKilo
    6. FPi
    7. GQwen Code
    8. HCrush
    9. ICline
    10. JPrime Agent
    11. KAider
    12. LGoose
    13. MHermes Agent
    14. NGrok Build
    15. ODeepSeek Harness
    16. POther coding tools and editor integrations
  8. 11Command line
    1. AInstall and update
    2. BSign in
    3. CStatus and models
    4. DCoding agents
  9. 12Console
    1. AAccount
    2. BWorkspaces
    3. CAPI keys
    4. DSettings
    5. EUsage
    6. FModels
    7. GPlayground
    8. HDeployments
  10. 13Billing
  11. 14Provider setup
    1. AWhen it works
    2. BCommands
    3. CSettings
    4. DProvider Console
    5. EEarnings
    6. FCrash reports
    7. GData centers
  12. 15Semantic routing
  13. 16Guardrails
  14. 17Knowledge

Reference

  1. 18API reference
    1. AAuthentication
    2. BChat completions
    3. CResponses
    4. DMessages
    5. ECount tokens
    6. FModels
    7. GPrices API
    8. HTasks
    9. IEmbeddings and rerank
    10. JImages
    11. KJobs
    12. LUploads
    13. MFiles
    14. NBatches
    15. OWebhooks
    16. PVector stores
    17. QDeployments
    18. RRequirements and headers
    19. SErrors
    20. TUsage
    21. UVersioning
  2. 19Operating guidance
    1. ALimits
    2. BErrors
    3. CRetries and billing outcomes
    4. DIdempotency
    5. EStreaming
    6. FData handling
    7. GCompatibility matrix
  3. 20Changelog
19
  1. Docs/
  2. Operating guidance/
  3. Data handling
(Detail F)
llms.txt

Data handling

On this sheet7 sections
  1. Your prompt data
  2. Your model weights
  3. Customer models: the import path
  4. What the gateway stores
  5. Routers and guardrails
  6. Pools and runtime access
  7. Regions
F
Fig. 1, sheet 19Operating guidance

Your prompt data

Policy

HIPS never trains on, retains or logs your prompt data. For anyone, on any plan: no prompt, file, result or response is used to train any model, and none is kept or written to a log. This includes requests processed by customer-owned models.

HIPS holds content only where you ask it to, to deliver what you asked for (batch files for thirty (30) days, task results for seven (7) days, uploads for twenty-four (24) hours, and Knowledge documents, files kept for later use and agents' own definitions and kept disks until you delete them, or thirty (30) days after their storage goes unpaid), keeps a prompt cache only in the GPU memory of its own servers, and on a provider's computer only in the engine's memory once it has given notice of it, re-runs a result to verify it only in memory, leaves it on a provider's computer only in the engine's memory until it is overwritten, and otherwise keeps content only where the law requires (Common Clauses Section 3.14).

The Common Clauses define what this covers: Prompt Data means all the content of a Request and of its response, whatever the interface, the Plan or the pool: Request Data; Batch Files; files uploaded to the Services; the results of task Requests and of jobs; documents added to a Workspace for the Services to search (Knowledge), and the passages the Services find in them; the inputs and the results of tools that HIPS runs for a Request; and anything kept for an agent. The Console will show this before you send work.

Your model weights

Policy

Customer-owned model weights will never be used to train other models on any plan.

Customer models: the import path

A customer model is imported from a model hub, never uploaded, and it is weight files only: no customer code runs anywhere. The import pins the revision, fetches only the files an engine reads, re-saves what passes its checks, copies it once into encrypted HIPS storage in the workspace's storage region, where its deployments run, and signs it with a HIPS key. Workers and devices load only that signed copy, never the hub. Import and run your model has the whole path.

A hub token for a private or gated repository is read-only, kept encrypted under a cloud key, shown by its last four characters, used only in the import job and never sent to a data-center worker, its RunPod pod or a device; it is deleted when you delete it. Access tokens has the rules.

What the gateway stores

WhatThe gateway
Request bodies, prompts, inputsNever stored. Read, validated, relayed to the worker, forgotten. Logged as sizes and counts, never as content.
Responses, outputsNever stored, except a task or job result on the job record until expires_at, so a client can retrieve it.
UploadsNever pass through the gateway (a presigned PUT). Stored encrypted in the HIPS account under the upload's retention.
The user fieldDiscarded.
The usage recordOperational metadata only: ids, timestamps, units, cost, outcome. No prompt, no input, no output.
  • What
    Request bodies, prompts, inputs
    The gateway
    Never stored. Read, validated, relayed to the worker, forgotten. Logged as sizes and counts, never as content.
  • What
    Responses, outputs
    The gateway
    Never stored, except a task or job result on the job record until expires_at, so a client can retrieve it.
  • What
    Uploads
    The gateway
    Never pass through the gateway (a presigned PUT). Stored encrypted in the HIPS account under the upload's retention.
  • What
    The user field
    The gateway
    Discarded.
  • What
    The usage record
    The gateway
    Operational metadata only: ids, timestamps, units, cost, outcome. No prompt, no input, no output.

Routers and guardrails

A request that names a router is read first by a decision service run outside HIPS: a masked excerpt of at most 4,096 bytes, the newest user message's text and then the system text, with every email address, phone number, Social Security number, card number, IBAN, IP address and secret replaced by a typed placeholder, for the one call. It is never stored, logged or sent to a model's server, and the usage record keeps the router, the preference and how it chose, never the task or the difficulty (Semantic routing).

A content moderation guardrail holds a matched value and a redacted original in the gateway's memory for the request alone: never stored, logged, returned or sent to a model's server, so a GPU receives only the redacted text. A check the decision service judges sends it only the parts the guardrail scans, after the patterns' redaction, for the one call (Guardrails), and a request that requires a region sends it nothing (A request that requires a region).

Pools and runtime access

In the community pool a job runs on a device that its owner controls physically. The provider app handles weights and inputs in memory and removes them after the job, and the software is signed; but a person with physical control of a machine can in principle read its memory, and HIPS cannot verify from outside that they did not. HIPS discloses this before a customer selects the pool. In the certified pool a job runs on data-center supply operated by an approved provider, with published execution protections and access conditions per provider Not decided yet.

Encryption in transit and at rest is part of the platform requirements. It protects data during transfer and storage, but does not by itself prevent a GPU operator from accessing weights or request data during execution. We will publish the execution protections and access conditions for each pool before customers deploy workloads to it.

Cross-pool fallback is off by default and requires explicit enablement. It never relaxes a region or data-use rule. Customer-owned models run in the certified pool only at launch.

Regions

A region is a requirement of a request, available on the Business plan only. What a region requirement covers (processing, logs, storage, backups, retries, verification samples, operations access, sub-processors) is open Not decided yet, and it is not offered until specified. The rule that holds regardless: no request silently crosses a required boundary. If no eligible supply exists in the required region, the gateway answers 503 overloaded_error with Retry-After, never a result from elsewhere.

Related sheets

  • Security and Data
  • Requirements and headers
Previous sheetStreaming
Next sheetCompatibility matrix
Sheet
19 of 20
Detail
F
Updated
2026-11-16
Sign in to the Console

On this sheet

  1. Your prompt data
  2. Your model weights
  3. Customer models: the import path
  4. What the gateway stores
  5. Routers and guardrails
  6. Pools and runtime access
  7. Regions

Product

  • Inference
  • Models
  • Deploy your model
  • Always-on GPU capacity
  • How it works
  • Use cases
  • Pricing

Developer

  • Docs
  • Quickstart
  • Migrate to HIPS
  • HIPS SDK
  • Batch jobs
  • API reference

Provider

  • Earn with your GPU
  • Register GPU interest
  • Download for Mac
  • Download for Windows
  • Download for Linux

Resource

  • Security and Data
  • Resources
  • llms.txt

Compare

  • OpenRouter alternatives
  • Together AI alternatives
  • Groq alternatives
  • All comparisons

Company

  • About
  • Contact
  • Affiliate program
  • HIPS Payments

Legal

  • Legal
  • Consumer Health Data Privacy Policy
  • Report content
  • Cookie preferences

Support

  • Contact HIPS
  • Looking for payments support?

HIPS. AI inference for builders.

hips.com

Hyperscale Inference Processing Systems OÜ