#3 Claude Opus 4.7
medium- Cost
- $0.059
- Time
- 26.8s
- Tokens
- 2,475 tok
Anthropic
Release: 2026-04-16
Tested on: 2026-04-16 15:59
anthropic/claude-opus-4.7::medium
~5T total (~500B active)MoEClosedEstimated
(medium)
(none)
Summary
Claude Opus 4.7 scores 9.2 on AI BENCHY and ranks #3. It has N/A reliability, a 88.9% pass rate, $0.447 total cost, and 3.53s average response time.
What makes Claude Opus 4.7 unique: It stands out most in Domain specific, where it ranks #2, while Anti-AI Tricks is its weakest area at #11. It is one of the highest-ranked models overall. It is notably fast compared with similar models.
Model facts
Researched on 2026-08-12
- Parameters
- ~5T total (~500B active)
- Architecture
- MoE
- Availability
- Closed
- License
- -
Best estimate from public evidence; the vendor did not disclose every value. External industry estimate; vendor has not disclosed the architecture.
9.2
Consistency
10.0
N/A
$0.447
Total Output Tokens
6,716
Total Input Tokens
0
Input Price
$5.000 / 1M
Output Price
$25.000 / 1M
Flaky tests
0
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-08-14 02:08 Re-test | 8.7 | 10.0 | $1.476 | Compare | |
| 2026-07-16 22:28 New test added | 8.7 | 10.0 | $1.477 | Compare | |
| 2026-06-04 13:44 New test added | 8.7 | 10.0 | $0.679 | Compare | |
| 2026-05-22 00:18 Re-test | 8.9 | 10.0 | $0.625 | Compare | |
| 2026-04-16 15:59 First recorded run | 9.2 | N/A | $0.447 | Current run |
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-04-16 15:59 · First recorded run | 54/54 attempts | 9.2 | 10.0 | N/A | 16/18 | 0 | 6,716 | 0 | $0.447 | 3.53s |
| 2026-07-16 22:28 · New test added | 66/66 attempts | 8.7 | 9.6 | 10.0 | 18/22 | 1 | 29,990 | 145,252 | $1.477 | 7.61s |
| Difference | — | +0.5 | +0.4 | -2 | -1 | -23274 | -145252 | -$1.030 | -4084ms |
Benchmark coverage differs: 54/54 attempts (Target: 3 repeats per test) versus 66/66 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.
These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.
Charts
Choose the first model, then click a second model to open a side-by-side page.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
Claude Opus 4.7mediumvsGLM 5.3 FlashmaxClaude Opus 4.7mediumvsMuse Spark 1.2highClaude Opus 4.7mediumvsGemini 3.6 FlashlowClaude Opus 4.7mediumvsMuse Spark 1.2mediumClaude Opus 4.7mediumvsClaude Opus 5lowClaude Opus 4.7mediumvsMuse Spark 1.1mediumClaude Opus 4.7mediumvsClaude Opus 4.8mediumClaude Opus 4.7mediumvsClaude Fable 5mediumClaude Opus 4.7mediumvsGemini 3.5 Flash LitehighClaude Opus 4.7mediumvsQwen3.6 Max Previewmedium
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 8.3 | 10.0 | |
| Coding | 10.0 | 10.0 | |
| Combined | 10.0 | 10.0 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 7.7 | 10.0 | |
| General Intelligence | 10.0 | 10.0 | |
| Instructions following | 10.0 | 10.0 | |
| Puzzle Solving | 10.0 | 10.0 | |
| Tool Calling | 10.0 | 10.0 |