The Open-Source AI Stack
RSS
All models

Models · claude

Claude Opus 4.5

Proprietary Anthropic · 2025-11-24 · Proprietary

Anthropic's first Opus to cross 80 percent on SWE-Bench Verified per the lab's own numbers, released November 24 2025 at a two-thirds price cut versus Opus 4.1 ($5 / $25 per Mtok). Added an effort parameter for adjustable reasoning intensity, with medium-effort runs matching Sonnet 4.5 using 76 percent fewer output tokens.

Cost

$5.00 / Mtok input
$25.00 / Mtok output

Anthropic API · as of 2026-05-21

source ↗

Speed

51.8 tok/sec output
1498 ms TTFT

· as of 2026-05-21

source ↗

Architecture

tokens in Embedding vocab not disclosed × N layers Architecture not disclosed (proprietary or undocumented) Output projection tokens out
Schema-generated from data/models.yaml. Every label is auditable against the model's sources.

Specs

Architecture
unknown
Total params
not disclosed
Active params
not disclosed
Context window
not verified
Attention
unknown
Position encoding
unknown
Post-training
rlhf, constitutional
OSI-approved
no
Data released
no
Training code
not released

Benchmarks

Each score carries the date it was published; we never infer or interpolate missing scores.

General reasoning

MMLU-Pro 88.9 as of 2026-05-21 source ↗
GPQA-Diamond 81.0 as of 2026-05-21 source ↗

Code

LiveCodeBench 73.8 as of 2026-05-21 source ↗

Math

AIME 2025 62.7 as of 2026-05-21 source ↗

Available quantizations

None. The weights are not distributed, so there are no public quantizations.

Notable innovations

  • · Adjustable effort parameter for reasoning intensity
  • · Two-thirds price cut versus Opus 4.1
  • · Plan Mode with clarifying questions in Claude Code

Lineage

Two-thirds price cut versus Opus 4.1; first Opus over 80% SWE-Bench Verified per Anthropic.

Derived from

Claude Opus 4.1 2025-08-05

Derivatives

Sources