Skip to content

Live catalogue

Published models and prices

Prices are in credits per million tokens, where one credit is one US dollar of vendor list-price usage. Credits are bought at 85% off list, so the effective rate you pay is a fraction of the struck-through number.

Public models with context window and per-million-token price in credits, vendor list price, capabilities, current availability
ModelContextList / 1MInput / 1MOutput / 1MCapabilitiesAvailability
claude-fable-5Claude Fable 51,000K$2.00 / $15.005.00 cr25.00 crStreaming · ToolsServingok
claude-opus-5Claude Opus 51,000K$5.00 / $25.005.10 cr25.10 crStreaming · ToolsServingok
claude-opus-4.6Claude Opus 4.61,000K$5.00 / $25.005.00 cr25.00 crStreaming · Tools · Vision · Prompt cacheServingok
claude-opus-4.8Claude Opus 4.8200K$5.00 / $25.005.00 cr25.00 crStreaming · Tools · Vision · Prompt cacheServingok
gpt-5.6-solGPT-5.6 Sol200K$5.00 / $25.005.00 cr25.00 crStreaming · Tools · VisionServingok

Why the list is short

A model appears here only after a live upstream channel has answered a real request for it. Vendors and aggregators routinely advertise models their endpoints reject, so we publish what we have verified rather than what we have been promised.

Prices change when we change them

This table is read from the same pricing rows the billing engine charges against, so there is no stale marketing copy to drift out of sync. Rates are reviewed when a vendor moves theirs.

Per model

Context, output ceiling and capabilities

Request a longer context or a higher output ceiling than a model supports and you get a 4xx describing the limit, not a truncated answer.

  • Claude Fable 5

    claude-fable-5
    Family
    other
    Context window
    1,000,000 tokens
    Max output
    1,000,000 tokens
    Streaming
    Yes
    Tool use
    Yes
    Vision
    No
    Prompt cache
    No

  • Claude Opus 5

    claude-opus-5
    Family
    other
    Context window
    1,000,000 tokens
    Max output
    1,000,000 tokens
    Streaming
    Yes
    Tool use
    Yes
    Vision
    No
    Prompt cache
    No

  • Claude Opus 4.6

    claude-opus-4.6
    Family
    claude
    Context window
    1,000,000 tokens
    Max output
    32,000 tokens
    Streaming
    Yes
    Tool use
    Yes
    Vision
    Yes
    Prompt cache
    Yes

  • Claude Opus 4.8

    claude-opus-4.8
    Family
    claude
    Context window
    200,000 tokens
    Max output
    32,000 tokens
    Streaming
    Yes
    Tool use
    Yes
    Vision
    Yes
    Prompt cache
    Yes

  • GPT-5.6 Sol

    gpt-5.6-sol
    Family
    gpt
    Context window
    200,000 tokens
    Max output
    200,000 tokens
    Streaming
    Yes
    Tool use
    Yes
    Vision
    Yes
    Prompt cache
    No

Programmatic

The same list over the API

Anything on this page is available from the gateway, so your tooling never has to scrape it.

bash
curl https://fuelofai.com/v1/models \
  -H "Authorization: Bearer $FUELOFAI_API_KEY"

Availability dots come from our own health records and are aggregated across every upstream route a model has. See the status page →

FuelOfAI is an independent API aggregator and is not affiliated with, endorsed by, or sponsored by Anthropic, OpenAI, or Google. All trademarks belong to their owners.