OpenAI model product
GPT-OSS-120B
GPT-OSS-120B High is an OpenAI language model variant designed for complex problems with high-effort thinking.
Updated Oct 23, 2026. Default version: GPT OSS 120B High
LLMBoard Score42.8GPT OSS 120B
Coverage60%7 benchmark families
Context window131.1KTokens
Official input priceN/AOfficial price unavailable
GPT-OSS-120B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-OSS-120B Benchmark Results
Benchmark scores for GPT OSS 120B.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkHealthBench Hard | Score30.00% | Rank06 | Participants11 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkHealthBench | Score57.60% | Rank07 | Participants14 | Percentile53.85% | EvidenceC | Evaluated |
| BenchmarkCodeForces | Score82.10% | Rank08 | Participants20 | Percentile63.16% | EvidenceC | Evaluated |
| BenchmarkAA AIME 2025 | Score93.44% | Rank11 | Participants96 | Percentile89.47% | EvidenceB | EvaluatedN/A |
| BenchmarkMMLU | Score90.00% | Rank12 | Participants103 | Percentile89.22% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Retail | Score67.80% | Rank15 | Participants25 | Percentile41.67% | EvidenceC | Evaluated |
| BenchmarkAA IFBench | Score68.98% | Rank51 | Participants168 | Percentile70.06% | EvidenceB | EvaluatedN/A |
| BenchmarkAA SciCode Subtasks | Score34.03% | Rank76 | Participants89 | Percentile14.77% | EvidenceB | EvaluatedN/A |
| BenchmarkHumanity's Last Exam | Score14.90% | Rank85 | Participants109 | Percentile22.22% | EvidenceC | Evaluated |
| BenchmarkAA HLE Text No Tools | Score19.60% | Rank90 | Participants200 | Percentile55.28% | EvidenceB | EvaluatedN/A |
| BenchmarkAA CritPt | Score1.14% | Rank94 | Participants200 | Percentile53.27% | EvidenceB | EvaluatedN/A |
| BenchmarkGPQA | Score80.10% | Rank103 | Participants256 | Percentile60.00% | EvidenceC | Evaluated |
| BenchmarkAA GPQA Diamond | Score78.18% | Rank106 | Participants199 | Percentile46.97% | EvidenceB | EvaluatedN/A |
| BenchmarkAA Omniscience Accuracy | Score21.78% | Rank112 | Participants201 | Percentile44.50% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Text Factuality | Score1,384.64 rating | Rank120 | Participants131 | Percentile8.46% | EvidenceA | Evaluated |
| BenchmarkAA LCR v1.1 | Score52.00% | Rank125 | Participants193 | Percentile35.42% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Text | Score1,365.55 rating | Rank136 | Participants218 | Percentile37.79% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,351.83 rating | Rank152 | Participants218 | Percentile30.41% | EvidenceA | Evaluated |
GPT-OSS-120B Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
GPT-OSS-120B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
GPT-OSS-120B Runtime Performance
Provider-specific output speed and catalog latency for GPT OSS 120B High. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderOpenAI | Output Speed100.00 tok/s | Catalog Latency6.50 s | Max Input131.1K | Max Output131.1K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-OSS-120B Specifications
Technical details for the model's default version.
- Version
- GPT OSS 120B High
- Released
- Aug 5, 2025
- Knowledge cutoff
- Unknown
- Parameters
- 116.8B
- Context window
- 131.1K
- Max output
- 131.1K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
GPT-OSS-120B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGPT OSS 120B | Released | LLMBoard42.76 | Parameters116.8B | Context131.1K | Max output131.1K | Open weightsYes | LicenseApache 2.0 |
| VersionGPT OSS 120B High | Released | LLMBoardN/A | Parameters116.8B | Context131.1K | Max output131.1K | Open weightsYes | LicenseApache 2.0 |
Models similar to GPT-OSS-120B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
o4 mini
OpenAI
OP
GPT-5-Mini
OpenAI
OP
GPT-5-Codex
OpenAI
OP
GPT-5.3
OpenAI
OP
GPT-OSS-20B
OpenAI
OP
GPT-5.4-nano
OpenAI
What is GPT-OSS-120B?
Key information about GPT-OSS-120B and its available data.
GPT-OSS-120B High is an OpenAI LLM variant that provides deeper analysis and more thorough responses than the base model. 1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization.
Data as of 2026-10-22.
FAQ
Common questions about GPT-OSS-120B.
When was GPT-OSS-120B released?
GPT-OSS-120B's default version was released on Aug 5, 2025.
How much does GPT-OSS-120B cost?
No official standard PAYG price is currently available for GPT-OSS-120B.
Who created GPT-OSS-120B?
GPT-OSS-120B was created by OpenAI.
What is the context window for GPT-OSS-120B?
The default version has a 131.1K token context window.
Is GPT-OSS-120B open weight?
Yes. The default version is marked as open weight under Apache 2.0.
How many API providers offer GPT-OSS-120B?
No provider offering is currently linked to the default version.