Parameters
Undisclosed
Context Length
200K
Modality
Multimodal
Architecture
Undisclosed
License
Proprietary
Release Date
16 Apr 2026
Knowledge Cutoff
-
API Pricing (per 1M)
Input: $5.00 · Output: $25.00
Rank
#15
| Benchmark | Score | Rank |
|---|---|---|
Graduate-Level QA | 0.942 | 🥉 3 |
General Text | high 1502 Standard 1494 | 4 7 |
Coding | max 0.82 | 7 |
Agent Arena | high 0.06 Standard 0.06 | 11 14 |
Mathematics | max 0.93 | 14 |
Data Analysis | max 0.78 | 14 |
General | max 0.77 | 17 |
LiveBench Average | max 0.77 | 17 |
Reasoning | max 0.87 | 20 |
Web Development | high 1556 Standard 1557 | 22 21 |
Coding Index | max 0.74 | 24 |
Agentic Coding | max 0.51 | 25 |
Intelligence Index | max 0.44 Standard 0.35 | 31 61 |
Agentic Index | max 0.40 | 35 |
Overall Rank
#15
Coding Rank
#13
Anthropic's most capable Claude 4.7 model, offering a notable improvement on Opus 4.6 in advanced software engineering and agentic tasks. Handles complex, long-running tasks with rigor and consistency, pays precise attention to instructions, and devises ways to verify its own outputs. Features substantially better vision (up to 2,576 pixels on the long edge — 3×+ the prior limit), improved multimodal understanding, and enhanced instruction following. Supports a new xhigh effort level between high and max for finer reasoning control. State-of-the-art on SWE-bench Verified and top performance across coding, finance, and document reasoning benchmarks. Available via API as claude-opus-4-7. Pricing: $5/M input tokens, $25/M output tokens.
Architecture specifications are undisclosed for proprietary models.
Claude 4.7 is Anthropic's latest generation of flagship models, offering notable improvements in advanced software engineering, agentic tasks, and instruction following. The family features substantially better vision (up to 2,576 pixels on the long edge), improved multimodal understanding, and a new xhigh effort level for finer reasoning control.
APX AI
Online