#164 Gemma 4 31B
medium- Cost
- $0.002
- Time
- 45.7s
- Tokens
- 2,696 tok
Google
Release: 2026-04-02
Tested on: 2026-08-14 02:55
google/gemma-4-31b-it::medium
30.7BDenseOpen source
(medium)
(none)
Summary
Gemma 4 31B scores 6.2 on AI BENCHY and ranks #164. It has 10.0 reliability, a 65.2% pass rate, $0.101 total cost, and 78.59s average response time.
What makes Gemma 4 31B unique: It stands out most in General Intelligence, where it ranks #1, while Tool Calling is its weakest area at #15.
Model facts
Researched on 2026-08-12
- Parameters
- 30.7B
- Architecture
- Dense
- Availability
- Open source
- License
- Apache-2.0
6.2
Consistency
8.7
10.0
Total Output Tokens
267,421
Total Input Tokens
94,969
Input Price
$0.100 / 1M
Output Price
$0.340 / 1M
Flaky tests
3
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 02:55 Re-test | 6.2 | 10.0 | $0.101 ↓ | Current run | |
| 2026-07-16 23:24 New test added | 6.3 | 10.0 | $0.163 ↑ | Compare | |
| 2026-06-04 14:27 New test added | 7.9 | 8.9 | $0.035 ↓ | Compare | |
| 2026-05-22 00:35 Suite changed | 8.0 | 6.7 | $0.029 | Compare | |
| 2026-04-14 00:56 First recorded run | 8.3 | N/A | $0.018 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-08-14 02:55 · Current run | 66/66 attempts | 6.2 | 8.7 | 10.0 | 13/22 | 3 | 267,421 | 94,969 | $0.101 | 78.59s |
| 2026-05-22 00:35 · Suite changed | 60/60 attempts | 8.0 | 9.2 | 6.7 | 14/20 | 2 | 69,291 | 0 | $0.029 | 35.51s |
| Difference | — | -1.8 | -0.5 | +3.3 | -1 | +1 | +198130 | +94969 | +$0.072 | +43082ms |
Benchmark coverage differs: 66/66 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.
Price History
Historical pricing data for this model from OpenRouter.
| Date | Input Price | Output Price |
|---|---|---|
| 2026-06-04 15:40 | $0.120 / 1M | $0.370 / 1M |
| 2026-06-05 23:32 | $0.120 / 1M | $0.360 / 1M |
| 2026-06-12 18:42 | $0.120 / 1M | $0.350 / 1M |
| 2026-07-14 10:48 | $0.060 / 1M | $0.350 / 1M |
| 2026-07-16 15:36 | $0.220 / 1M | $0.550 / 1M |
| 2026-07-25 03:22 | $0.140 / 1M | $0.400 / 1M |
| 2026-08-14 12:51 | $0.100 / 1M | $0.340 / 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
Gemma 4 31BmediumFree AvailablevsSeed-2.0-LitenoneGemma 4 31BmediumFree AvailablevsGPT-5.6 LunalowGemma 4 31BmediumFree AvailablevsRing-2.6-1TmediumGemma 4 31BmediumFree AvailablevsClaude Sonnet 5noneGemma 4 31BmediumFree AvailablevsLing-3.0-flashhighGemma 4 31BmediumFree AvailablevsGemini 3.1 Flash LiteminimalGemma 4 31BmediumFree AvailablevsMiMo-V2.5mediumGemma 4 31BmediumFree AvailablevsQwen3.5-35B-A3BmediumGemma 4 31BmediumFree AvailablevsLongCat 2.0noneGemma 4 31BmediumFree Availablevsgpt-oss-120bmedium
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 10.0 | 10.0 | |
| Coding | 4.3 | 5.8 | |
| Combined | 2.9 | 5.8 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 5.3 | 7.2 | |
| General Intelligence | 10.0 | 10.0 | |
| Instructions following | 10.0 | 10.0 | |
| Puzzle Solving | 9.9 | 10.0 | |
| Tool Calling | 3.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |