# GPT-OSS-120B finance AI benchmark results

Every finance benchmark score for GPT-OSS-120B by OpenAI: 2 benchmarks, each with its rank, metric and source date. Scores are never combined across benchmarks.

## Board standings

GPT-OSS-120B is first on 0 benchmarks, top 3 on 1 benchmark, scored on 2 benchmarks. Places on each benchmark's main board with 2 or more models; scores are never combined. https://benchmarks.finance/method/#standings
- CAIA (Crypto AI Agent Benchmark): #2 of 10

## Scores

| Benchmark | Rank | Score | Metric | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [CAIA (Crypto AI Agent Benchmark)](https://benchmarks.finance/benchmarks/caia/) | 2 of 10 | 62.9 | accuracy % (majority vote, 5 runs) | Paper Tables 2-3 | 17 Jan 2026 |
| [PolyBench](https://benchmarks.finance/benchmarks/polybench/) | 6 of 7 | -17.8 | Confidence-Weighted Return (CWR) % | Paper Table 5 | 3 Apr 2026 |

Source: https://benchmarks.finance/models/gpt-oss-120b/

Created and maintained by Moneda.