Today's price · per 1M tokens
Input
$5.00
per 1M tokens
Output
$25.00
per 1M tokens
Blended
$15.00
avg of input & output $/1M
Cached input
$0.500
10% of input — prompt caching
Source: official Anthropic pricing · read Jun 25, 2026
MODELPRICEWATCH.COM · 2026-08-01
Overview
Anthropic's Opus 4.6 — a frontier model with class-leading coding (~80.8% SWE-bench Verified) and standout legal/enterprise reasoning (top BigLaw Bench score). 1M context, prompt caching up to 90% off. $5/$25 per 1M tokens.
Capabilities
struck through = not supportedInput 2/5
Text ✓
Image ✓
Audio
Video
PDF
Output 1/5
Text ✓
Image
Audio
Video
Embedding
Features 4/9
Prompt caching ✓
Reasoning ✓
Coding
Fast inference
Long context ✓
Open weights
Multimodal ✓
Web search
Realtime
Benchmark performance
accuracy % · higher is betterAvg benchmark score76.7%
Perf per $/Mtok5.1
GPQA Diamond
91.3%
Terminal-Bench 2.1
65.4%
LiveCodeBench
76%
MCP Atlas
76.8%
Humanity's Last Exam
40%
ARC-AGI 2
68.8%
AIME 2025
99.8%
MMMLU
91.1%
HumanEval
80.8%
Independent composite scores
- LMSYS Chatbot Arena: 1497.2 ELO (Human preference)
How it stacks up
- Ranks #20 of 68 benchmarked models by average score
- Ranks #22 of 133 comparably-measured models by percentile score, across 10 independent measurements
- Ranks #116 of 133 comparably-tested models by normalized performance per dollar
- Strongest at MMMLU — 91.1%, #1 of 33
67 tokens/sec output 1.6s latency to first token (TTFT)
Source: Vellum LLM Leaderboard · updated Aug 1, 2026 · See full rankings →
Specifications
- Provider
- Anthropic
- Context window
- 1M tokens
- Modality
- text, image
- Parameters
- Proprietary
- Open source
- No — proprietary
- Released
- Feb 5, 2026
- Status
- Current
- Last updated
- Jun 25, 2026
- Tags
Availability verified: Jul 25, 2026 — listed on Anthropic's own page