#13 Claude Opus 4.8
medium- Cost
- $0.057
- Time
- 23.1s
- Tokens
- 2,412 tok
Anthropic
Release: 2026-05-28
Tested on: 2026-07-16 22:32
anthropic/claude-opus-4.8::medium
~5T total (~500B active)MoEClosedEstimated
Summary
Claude Opus 4.8 scores 8.8 on AI BENCHY and ranks #13. It has 10.0 reliability, a 84.9% pass rate, $1.931 total cost, and 12.49s average response time.
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.
8.8
Consistency
9.6
10.0
$1.931
Total Output Tokens
49,841
Total Input Tokens
138,451
Input Price
$5.000 / 1M
Output Price
$25.000 / 1M
Flaky tests
1
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 02:10 Re-test | 8.8 | 10.0 | $1.983 | Compare | |
| 2026-07-16 22:32 New test added | 8.8 | 10.0 | $1.931 | Current run | |
| 2026-06-30 22:01 Re-test | 8.8 | 10.0 | $1.107 | Compare | |
| 2026-06-04 13:45 New test added | 8.7 | 10.0 | $1.107 | Compare | |
| 2026-05-28 17:34 Initial run | 8.7 | 10.0 | $1.006 | Compare |
This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-07-16 22:32 · New test added | 66/66 attempts | 8.8 | 9.6 | 10.0 | 18/22 | 1 | 49,841 | 138,451 | $1.931 | 12.49s |
| 2026-06-04 13:45 · New test added | 63/63 attempts | 8.7 | 9.6 | 10.0 | 17/21 | 1 | 32,396 | 61,007 | $1.107 | 9.66s |
| Difference | — | +0.1 | 0.0 | 0.0 | +1 | 0 | +17445 | +77444 | +$0.824 | +2826ms |
Benchmark coverage differs: 66/66 attempts (Target: 3 repeats per test) versus 63/63 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.
Price History
Historical pricing data for this model from OpenRouter.
| Date | Input Price | Output Price |
|---|---|---|
| 2026-06-04 15:40 | $5.000 / 1M | $25.000 / 1M |
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 4.8mediumvsGemini 3.5 Flash LitehighClaude Opus 4.8mediumvsClaude Opus 5lowClaude Opus 4.8mediumvsGemini 3.5 FlashlowClaude Opus 4.8mediumvsGemini 3.6 FlashlowClaude Opus 4.8mediumvsSeed 2.1 TurbohighClaude Opus 4.8mediumvsGLM 5.3 FlashmaxClaude Opus 4.8mediumvsGPT-5.3-CodexmediumClaude Opus 4.8mediumvsClaude Opus 4.7mediumClaude Opus 4.8mediumvsGrok 4.5highClaude Opus 4.8mediumvsMuse Spark 1.2high
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 10.0 | 10.0 | |
| Coding | 10.0 | 10.0 | |
| Combined | 9.9 | 10.0 | |
| Data parsing and extraction | 7.1 | 5.6 | |
| Domain specific | 5.3 | 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 | |
| Trivia | 3.0 | 10.0 |