#112 gpt-oss-120b
medium- Cost
- $0.001
- Time
- 26.7s
- Tokens
- 555 tok
OpenAI
Release: 2025-08-05
Tested on: 2026-05-22 00:18
openai/gpt-oss-120b::medium
117B total (5.1B active)MoEOpen source
(medium)
(none)
Summary
gpt-oss-120b scores 5.6 on AI BENCHY and ranks #112. It has 10.0 reliability, a 46.7% pass rate, $0.012 total cost, and 20.89s average response time.
What makes gpt-oss-120b unique: It stands out most in Combined, where it ranks #1, while Anti-AI Tricks is its weakest area at #18. Its total benchmark cost is unusually low for its score range.
Model facts
Researched on 2026-08-12
- Parameters
- 117B total (5.1B active)
- Architecture
- MoE
- Availability
- Open source
- License
- Apache-2.0
5.6
Consistency
7.5
10.0
$0.012
Total Output Tokens
63,919
Total Input Tokens
0
Input Price
$0.000 / 1M
Output Price
$0.000 / 1M
Flaky tests
6
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 | 6.1 | 10.0 | $0.019 ↓ | Compare | |
| 2026-07-16 22:24 New test added | 6.1 | 10.0 | $0.019 ↓ | Compare | |
| 2026-06-04 13:44 New test added | 6.1 | 10.0 | $0.013 ↓ | Compare | |
| 2026-05-22 00:18 Suite changed | 5.6 | 10.0 | $0.012 | Current run | |
| 2026-04-21 12:42 First recorded run | 5.8 | N/A | $0.011 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-22 00:18 · Suite changed | 60/60 attempts | 5.6 | 7.5 | 10.0 | 7/20 | 6 | 63,919 | 0 | $0.012 | 20.89s |
| 2026-06-04 13:44 · New test added | 63/63 attempts | 6.1 | 8.0 | 10.0 | 9/21 | 5 | 70,246 | 39,084 | $0.013 | 22.28s |
| Difference | — | -0.5 | -0.5 | 0.0 | -2 | +1 | -6327 | -39084 | -$0.001 | -1395ms |
Benchmark coverage differs: 60/60 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.
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
gpt-oss-120bmediumFree AvailablevsQwen3.5-35B-A3Bmediumgpt-oss-120bmediumFree AvailablevsQwen3.7 Flashnonegpt-oss-120bmediumFree AvailablevsGemini 3.1 Flash Liteminimalgpt-oss-120bmediumFree AvailablevsGemini 3.1 Flash Litenonegpt-oss-120bmediumFree AvailablevsClaude Sonnet 5nonegpt-oss-120bmediumFree AvailablevsQwen3.5-Flashmediumgpt-oss-120bmediumFree AvailablevsGPT-5.6 Lunalowgpt-oss-120bmediumFree AvailablevsInklinglowgpt-oss-120bmediumFree AvailablevsGemma 4 31BmediumFree Availablegpt-oss-120bmediumFree AvailablevsTrinity Large Thinkingmedium
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 6.7 | 9.9 | |
| Coding | 3.9 | 5.6 | |
| Combined | 10.0 | 10.0 | |
| Data parsing and extraction | 6.4 | 5.9 | |
| Domain specific | 2.9 | 4.4 | |
| General Intelligence | 4.3 | 10.0 | |
| Instructions following | 9.9 | 10.0 | |
| Puzzle Solving | 3.2 | 4.7 | |
| Tool Calling | 9.8 | 10.0 | |
| Trivia | 3.0 | 10.0 |