/

/

Anthropic

MODEL PROVIDER

Anthropic

Anthropic’s Claude models, called directly through the Anthropic API. Includes the Opus, Sonnet, and Haiku tiers, with extended thinking on recent models.

Anthropic has 14 models available in SteelEngine with pricing, context window, and capability details.

MODELS

14

DEFAULT MODEL

Claude Opus 5

PRICING

USD PER 1M TOKENS

MODELS

Anthropic models in SteelEngine.

Recommended models come first. Open any model for its full price sheet, limits, and supported capabilities.

Anthropic

RECOMMENDED

Claude Opus 5.5

Mid-priced reasoning model for everyday agent steps that need deliberate thinking. A recommended default among SteelEngine’s Anthropic models.

IN / 1M

$4

OUT / 1M

$20

CONTEXT

1M

Anthropic

RECOMMENDED

Claude Sonnet 5.5

Lower-priced reasoning model for high-volume steps that still benefit from thinking. A recommended default among SteelEngine’s Anthropic models.

IN / 1M

$2

OUT / 1M

$10

CONTEXT

1M

Anthropic

RECOMMENDED

Claude Haiku 5.5

The lowest-cost model in the Anthropic lineup, for high-volume steps that still benefit from thinking. A recommended default among SteelEngine’s Anthropic models.

IN / 1M

$0.1

OUT / 1M

$0.5

CONTEXT

1M

Anthropic

RECOMMENDED

Claude Opus 5

Higher-priced reasoning model for complex planning and multi-step agent work. A recommended default among SteelEngine’s Anthropic models. Released Jul 24, 2026.

IN / 1M

$5

OUT / 1M

$25

CONTEXT

1M

Anthropic

RECOMMENDED

Claude Opus 4.8

Higher-priced reasoning model for complex planning and multi-step agent work. A recommended default among SteelEngine’s Anthropic models. Released May 28, 2026.

IN / 1M

$5

OUT / 1M

$25

CONTEXT

1M

Anthropic

RECOMMENDED

Claude Opus 4.7

Higher-priced reasoning model for complex planning and multi-step agent work. A recommended default among SteelEngine’s Anthropic models. Released Apr 16, 2026.

IN / 1M

$5

OUT / 1M

$25

CONTEXT

1M

Anthropic

Claude Fable 5.1

Among the highest-priced models in the Anthropic lineup, for the hardest reasoning and planning work. Reads up to 1M tokens of context in one call and writes up to 128k tokens. Released Sep 1, 2026.

IN / 1M

$10

OUT / 1M

$50

CONTEXT

1M

Anthropic

Claude Fable 5

Among the highest-priced models in the Anthropic lineup, for the hardest reasoning and planning work. Reads up to 1M tokens of context in one call and writes up to 128k tokens. Released Jun 9, 2026.

IN / 1M

$10

OUT / 1M

$50

CONTEXT

1M

Anthropic

Claude Sonnet 5

Lower-priced reasoning model for high-volume steps that still benefit from thinking. Reads up to 1M tokens of context in one call and writes up to 128k tokens. Released Jun 9, 2026.

IN / 1M

$2

OUT / 1M

$10

CONTEXT

1M

Anthropic

Claude Opus 4.6

Higher-priced reasoning model for complex planning and multi-step agent work. Reads up to 1M tokens of context in one call and writes up to 128k tokens.

IN / 1M

$5

OUT / 1M

$25

CONTEXT

1M

Anthropic

Claude Sonnet 4.6

Lower-priced reasoning model for high-volume steps that still benefit from thinking. Reads up to 1M tokens of context in one call and writes up to 128k tokens. Sonnet is Anthropic’s mid-size Claude tier.

IN / 1M

$3

OUT / 1M

$15

CONTEXT

1M

Anthropic

Claude Opus 4.5

Higher-priced reasoning model for complex planning and multi-step agent work. Handles 200k tokens of context and writes up to 64k tokens.

IN / 1M

$5

OUT / 1M

$25

CONTEXT

200k

Anthropic

Claude Sonnet 4.5

Lower-priced reasoning model for high-volume steps that still benefit from thinking. Handles 200k tokens of context and writes up to 64k tokens.

IN / 1M

$3

OUT / 1M

$15

CONTEXT

200k

Anthropic

Claude Haiku 4.5

Fast, lower-priced reasoning model for high-volume steps that still benefit from thinking. Handles 200k tokens of context and writes up to 64k tokens.

IN / 1M

$1

OUT / 1M

$5

CONTEXT

200k

BUILD WITH STEELENGINE

Put Anthropic models to work.

Use these models in deterministic workflow steps or production agents, with approvals, guardrails, and a full audit trail around every call.