PACINFRAX · PRODUCT

Compare model references and indicative costs.

Search the shared model and service reference book, then estimate text costs locally.

Your decision

Compare exact model identifiers and units without invoking a model.

  • Check the model passport for its source, reference date and unresolved license/service conditions.
  • Estimate input and output tokens separately; service charges and dedicated compute are separate units.
  • The seven-entry seed is partial. A record is not proof of an active endpoint or approved commercial rate.

Entries are non-callable and non-billable. Estimates are not quotes; no request, key, order or payment is submitted.

COMPARE YOUR OPTIONS

Models and service price units.

Price version pacinfrax-draft-2026-10-03-v1. This is a partial reference catalog; entries cannot receive traffic or create billable usage.

Catalog terms only—do not enter names or customer identifiers. Search is not submitted or persisted.

Category

Showing 7 of 7 reference entries

Text model referenceIndicative

GPT-OSS 120B

openai/gpt-oss-120b

Planning unit
USD / 1M input or output tokens
Indicative input rate
$0.13875
Indicative output rate
$0.555
License / cost validation
Pending / pending
Reference checked
2026-10-03
Reference source (external)

Inspect GPT-OSS 120B model passport →

Non-callable · Non-billable · Availability not established

Text model referenceIndicative

Qwen3.5 9B

Qwen/Qwen3.5-9B

Planning unit
USD / 1M input or output tokens
Indicative input rate
$0.15725
Indicative output rate
$0.23125
License / cost validation
Pending / pending
Reference checked
2026-10-03
Reference source (external)

Inspect Qwen3.5 9B model passport →

Non-callable · Non-billable · Availability not established

Service referenceIndicative

Shared Filesystem

shared-filesystem

Planning unit
USD / GiB-month
Indicative service rate
$0.148
License / cost validation
Pending / pending
Reference checked
2026-10-03
Reference source (external)

Non-callable · Non-billable · Availability not established

Service referenceIndicative

Code Sandbox — CPU

code-sandbox

Planning unit
USD / vCPU-hour
Indicative service rate
$0.041255
License / cost validation
Pending / pending
Reference checked
2026-10-03
Reference source (external)

CPU and RAM are separate pricing components, not a complete sandbox quote.

Non-callable · Non-billable · Availability not established

Service referenceIndicative

Code Sandbox — RAM

code-sandbox

Planning unit
USD / GiB RAM-hour
Indicative service rate
$0.0137825
License / cost validation
Pending / pending
Reference checked
2026-10-03
Reference source (external)

CPU and RAM are separate pricing components, not a complete sandbox quote.

Non-callable · Non-billable · Availability not established

Service referenceIndicative

Code Interpreter

code-interpreter

Planning unit
USD / session (60 minutes)
Indicative service rate
$0.02775
License / cost validation
Pending / pending
Reference checked
2026-10-03
Reference source (external)

Non-callable · Non-billable · Availability not established

COST CALCULATOR

Estimate a text request—without making one.

Choose a text-model reference and whole token counts. This calculation uses indicative prices on your device; it does not invoke a model, check availability, create a quote, or write usage.

  • No API key or prompt required
  • No data sent or persisted
  • Maximum 1,000,000,000 tokens per field

Reference IDs are not PacinfraX callable model IDs.

Whole number from 0 to 1,000,000,000.

Whole number from 0 to 1,000,000,000.

Planning result

No current estimate

Choose a model and token counts, then calculate. Changing any input clears the previous result.

Review detailed reference pricing

Continue your decision