OpenAI
Release: 2025-08-05
Tested on: 2026-10-01 05:17
openai/gpt-oss-120b::none
117B total (5.1B active)MoEOpen source
(medium)
(none)
Summary
gpt-oss-120b scores 4.2 on AI BENCHY and ranks #361. It has 10.0 reliability, a 34.8% pass rate, $0.016 total cost, and 28.63s average response time.
What makes gpt-oss-120b unique: It stands out most in Agentic, where it ranks #1, while Coding is its weakest area at #10.
Model facts
Researched on 2026-08-12
- Parameters
- 117B total (5.1B active)
- Architecture
- MoE
- Availability
- Open source
- License
- Apache-2.0
4.2
Consistency
7.6
10.0
Total Output Tokens
62,213
Total Input Tokens
131,652
Input Price
$0.037 / 1M
Output Price
$0.170 / 1M
Cache Read Price
N/A
Cache Write Price
N/A
Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.
Wrong Tests: 14
Attempt pass rate: 34.8%
Benchmark coverage: 60/69 attempts (Target: 3 repeats per test)
Flaky tests
3
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-10-01 05:17 Suite changed | 4.2 | 10.0 | $0.016 ↓ | Current run | |
| 2026-04-21 12:42 First recorded run | 5.2 | N/A | $0.009 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-10-01 05:17 · Current run | 60/69 attempts | 4.2 | 7.6 | 10.0 | 6/20 | 3 | 62,213 | 131,652 | $0.016 | 28.63s |
| 2026-04-21 12:42 · First recorded run | 54/54 attempts | 5.2 | 7.9 | N/A | 4/18 | 5 | 44,652 | 0 | $0.009 | 11.96s |
| Difference | — | -1.0 | -0.3 | +2 | -2 | +17561 | +131652 | +$0.007 | +16672ms |
Benchmark coverage differs: 60/69 attempts (Target: 3 repeats per test) versus 54/54 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 | Cache Read Price | Cache Write Price |
|---|---|---|---|---|
| 2026-06-04 15:40 | $0.039 / 1M | $0.180 / 1M | N/A | N/A |
| 2026-06-29 01:00 | $0.030 / 1M | $0.150 / 1M | N/A | N/A |
| 2026-07-09 20:59 | $0.036 / 1M | $0.180 / 1M | N/A | N/A |
| 2026-07-16 15:36 | $0.037 / 1M | $0.170 / 1M | N/A | N/A |
| 2026-08-12 11:59 | $0.030 / 1M | $0.170 / 1M | N/A | N/A |
Charts
Choose the first model, then click a second model to open a side-by-side page.
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
gpt-oss-120bnonevsGLM 4.7 Flashmediumgpt-oss-120bnonevsMercury 2.5nonegpt-oss-120bnonevsMiniMax M2.5mediumgpt-oss-120bnonevsQwen3.5-9Bmediumgpt-oss-120bnonevsMercury 2nonegpt-oss-120bnonevsLing 3.0 Tinydefaultgpt-oss-120bnonevsCobuddymediumFree Availablegpt-oss-120bnonevsGranite 4.1 8bdefaultgpt-oss-120bnonevsGranite 4.2 8Bnonegpt-oss-120bnonevsNemotron 3.5 LightningnoneFree Available
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Agentic | 6.1 | 3.1 | |
| Anti-AI Tricks | 6.5 | 10.0 | |
| Coding | 1.5 | 0.4 | |
| Combined | 1.5 | 5.0 | |
| Data parsing and extraction | 6.5 | 10.0 | |
| Domain specific | 3.0 | 10.0 | |
| General Intelligence | 4.8 | 10.0 | |
| Instructions following | 9.8 | 10.0 | |
| Puzzle Solving | 6.0 | 7.2 | |
| Tool Calling | 3.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |