#2 Gemini 3.5 Flash
low- Cost
- $0.068
- Time
- 39.1s
- Tokens
- 7,588 tok
Google
Release: 2026-05-19
Tested on: 2026-05-19 23:35
google/gemini-3.5-flash::low
~500B total (~20B active)MoEClosedEstimated
Summary
Gemini 3.5 Flash scores 9.6 on AI BENCHY and ranks #2. It has 10.0 reliability, a 94.7% pass rate, $0.245 total cost, and 2.84s average response time.
What makes Gemini 3.5 Flash unique: It stands out most in Domain specific, where it ranks #2, while Instructions following is its weakest area at #10. It is one of the highest-ranked models overall. Its total benchmark cost is unusually low for its score range.
Model facts
Researched on 2026-08-12
- Parameters
- ~500B total (~20B active)
- Architecture
- MoE
- Availability
- Closed
- License
- -
Best estimate from public evidence; the vendor did not disclose every value. Google does not disclose parameter counts.
9.6
Consistency
10.0
10.0
$0.245
Total Output Tokens
22,248
Total Input Tokens
0
Input Price
$1.500 / 1M
Output Price
$9.000 / 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 02:46 Re-test | 8.8 | 10.0 | $0.449 | Compare | |
| 2026-07-16 22:57 New test added | 8.9 | 10.0 | $0.433 | Compare | |
| 2026-06-04 14:02 New test added | 9.4 | 10.0 | $0.349 | Compare | |
| 2026-05-22 00:28 Suite changed | 9.3 | 10.0 | $0.285 | Compare | |
| 2026-05-19 23:35 Initial run | 9.6 | 10.0 | $0.245 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-19 23:35 · Initial run | 57/57 attempts | 9.6 | 10.0 | 10.0 | 18/19 | 0 | 22,248 | 0 | $0.245 | 2.84s |
| 2026-05-22 00:28 · Suite changed | 60/60 attempts | 9.3 | 10.0 | 10.0 | 18/20 | 0 | 25,965 | 0 | $0.285 | 2.98s |
| Difference | — | +0.3 | 0.0 | 0.0 | 0 | 0 | -3717 | 0 | -$0.040 | -137ms |
Benchmark coverage differs: 57/57 attempts (Target: 3 repeats per test) versus 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.
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
Gemini 3.5 FlashlowvsSeed 2.1 TurbohighGemini 3.5 FlashlowvsGemini 3.5 Flash LitehighGemini 3.5 FlashlowvsGPT-5.3-CodexmediumGemini 3.5 FlashlowvsClaude Opus 4.8mediumGemini 3.5 FlashlowvsGrok 4.5highGemini 3.5 FlashlowvsClaude Opus 5lowGemini 3.5 FlashlowvsQwen3.8 MaxmediumGemini 3.5 FlashlowvsGemini 3.6 FlashlowGemini 3.5 FlashlowvsGPT-5.5mediumGemini 3.5 FlashlowvsClaude Opus 4.7medium
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 10.0 | 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 | 9.9 | 10.0 | |
| Puzzle Solving | 10.0 | 10.0 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 10.0 | 10.0 |