Summary
Ling-2.6-flash scores 4.3 on AI BENCHY and ranks #243. It has N/A reliability, a 28.8% pass rate, $0.000 total cost, and 1.27s average response time.
What makes Ling-2.6-flash unique: It stands out most in Instructions following, where it ranks #1, while Combined is its weakest area at #8. Its total benchmark cost is unusually low for its score range.
Archived model: this model is no longer updated or tested on new tests.
Identity note
Elephant Alpha was the stealth version of Ling-2.6-flash.
This page currently reflects the benchmark data collected under Elephant Alpha.
Researched on 2026-08-12
4.3
Consistency
9.2
N/A
$0.000
Total Output Tokens
2,596
Total Input Tokens
33,744
Input Price
$0.000 / 1M
Output Price
$0.000 / 1M
Flaky tests
1
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-06-04 14:27 New test added | 4.3 | N/A | $0.000 | Current run | |
| 2026-05-22 00:40 Suite changed | 5.4 | N/A | $0.000 | Compare | |
| 2026-04-22 12:55 First recorded run | 5.2 | N/A | $0.000 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-06-04 14:27 · Current run | 21/22 tests · 63/66 attempts | 4.3 | 9.2 | N/A | 6/21 | 1 | 2,596 | 33,744 | $0.000 | 1.27s |
| 2026-05-22 00:40 · Suite changed | 20/20 tests · 60/60 attempts | 5.4 | 9.6 | N/A | 6/19 | 1 | 2,596 | 0 | $0.000 | 1.27s |
| Difference | — | -1.1 | -0.4 | -1 | 0 | 0 | +33744 | $0.000 | 0ms |
Benchmark coverage differs: 21/21 tests · 63/63 attempts (Target: 3 repeats per test) versus 20/20 tests · 60/60 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.
Choose the first model, then click a second model to open a side-by-side page.
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 6.6 | 10.0 | |
| Coding | 3.7 | 7.8 | |
| Combined | 1.5 | 5.0 | |
| Data parsing and extraction | 6.5 | 10.0 | |
| Domain specific | 3.0 | 10.0 | |
| General Intelligence | 4.3 | 10.0 | |
| Instructions following | 9.8 | 10.0 | |
| Puzzle Solving | 5.3 | 10.0 | |
| Tool Calling | 3.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |