GPT-4o
GPT-4o is OpenAI’s multimodal all-around model with native support for text, image, and audio inputs. It operates with a context window of 128,000 tokens, is available exclusively via the OpenAI API, and targets a broad range of productive applications — from analysis and coding to natural conversation.
- Proprietary
- Frontier
- API
- Text
- Vision
- Audio
- Instruction-Tuned
- Real-Time
Sovereign Risk: MEDIUM OpenAI is a US company and subject to the CLOUD Act. When using the API, input data leaves the local network — government access to processed data is legally possible.
Key metrics
Score · Latency · Cost · Quality
- Total Score Silver
- 67.99
- Routine
- 41.05
- Reasoning
- 26.94
- LLM Judge Avg
- 3.49 / 5
- 100 Coverage
- Avg Task Duration
- 4.39s
- Real-Time
- Token Rate
- 150.4tok/s
- Output Rate
- P95 Latency
- 9.58s
- Top 5 %
- Total Tokens
- 47900
- Output Volume
- Cost per 1K
- $0.01
- USD / 1K Requests
- Benchmark Cost
- $0.48
- Total · 47900 tok
Benchmark modules
10 modules · weighted · vs. model median & top performer
GPT-4o
Best model
Ø All models
Code Quality
70.1
CLI Benchmark
81.68
Logical Reasoning
66.04
UX Writing
62.55
Documentation
58.47
Content Transform.
78.3
Cultural Intelligence
66.8
Synthesis Quality
48.33
Tool Execution
86.67
ToolUse Score
66.83
Benchmark Cost
$0.48
Token efficiency & latency
Consumption per module vs. model median
Token consumption per module
Performance profile
Key metrics – legend
Score
- Total Score
- Overall score derived from Routine and Reasoning scores. Basis for the fleet rank and tier assignment (Gold / Silver / Bronze).
- Routine
- Score for standard tasks: code generation, text transformation, structured outputs. Weighted by task complexity.
- Reasoning
- Score for analytical and logical tasks: multi-step problem solving, mathematical reasoning, structured argumentation.
- Rank
- Position in the fleet comparison, sorted by Total Score. Refers to the snapshot at the time of testing.
LLM Judge
- LLM Judge Ø
- Average quality rating by an independent judge model on a scale of 1–5.
- Coverage
- Share of tasks for which a judge rating is available. 100% means all responses were evaluated.
Latency & speed
- Avg. Task Duration
- Average time per task from the first to the last token of the response, measured across all benchmark tasks.
- Token Rate
- Output speed in tokens per second. Relevant for interactive applications and real-time streaming.
- P95 Latency
- 95th percentile of response times — shows outliers. Lower values mean fewer long wait times in edge cases.
Cost
- Cost per 1K
- API price per 1,000 output tokens in US dollars. Basis for cost calculation in production use.
- Benchmark Cost
- Total cost for the complete benchmark run of this model.