Data points: 120
Model compare
The readout for GPT-5 and GPT-5.4 Mini, before the detailed comparison sheet.
Weighted outcome: GPT-5.4 Mini. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
Decision read
GPT-5.4 Mini
GPT-5.4 Mini has the higher weighted result; Model A / B score 34.7 to 65.3.
Evidence depth
120 data points
Includes 22 benchmark rows, 2 audit samples, and 7 provider examples.
Selection signal
Start with GPT-5.4 Mini
The charts below split 31 high-signal samples across speed, scores, and audit health.
Switch either side of this report to compare another model with the same LMSpeed data pipeline.
Select a different model to open a new comparison URL.
This report only uses LMSpeed data for GPT-5 and GPT-5.4 Mini: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
| Model compare | GPT-5 | GPT-5.4 Mini |
|---|---|---|
| Overall leader | Contender | Leading |
| Weighted overall score | 34.7 pts | 65.3 pts |
| Benchmark category leads | 2 categories | 4 categories |
| Operational advantages | Cheapest input price, Free providers | Average speed, First-token latency, Provider coverage |
| Context window | 400K tokens | 400K tokens |
| Max output | 128K tokens | 128K tokens |
| Modalities | Input TextImageFile Output Text | Input |
The overall result weights benchmark capability categories at 80% and price, API speed/latency, and availability at 20%. Recent test volume does not affect the winner, and missing benchmark categories are excluded.
| Model compare | GPT-5 | GPT-5.4 Mini |
|---|---|---|
| Developer | OpenAI | OpenAI |
| Released | Aug 2025 | Mar 2026 |
| Parameters | No data | No data |
| Tokenizer | GPT | GPT |
| Knowledge cutoff | 2024-09-30 | 2025-08-31 |
| OpenRouter ID | openai/gpt-5 | openai/gpt-5.4-mini |
| References | No data | No data |
This report only uses LMSpeed data for GPT-5 and GPT-5.4 Mini: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
GPT-5
GPT-5 is stronger in benchmark categories (Knowledge, Math) and operational dimensions (Cheapest input price, Free providers).
GPT-5.4 Mini
GPT-5.4 Mini is stronger in benchmark categories (Agents, Coding, Reasoning, Instruction following) and operational dimensions (Average speed, First-token latency, Provider coverage).
Third-party benchmark profile synced into LMSpeed; only metrics available for both models are shown.
Compare benchmark category scores on a 0-100 scale. Select a category to inspect the gap.
Avg. score
GPT-5
51.9
Avg. score
GPT-5.4 Mini
51.9
GPT-5.4 Mini leads by 5.6
GPT-5.4 Mini leads by 8.8
GPT-5.4 Mini leads by 4.5
GPT-5 leads by 2.5
GPT-5 leads by 11.9
No data
GPT-5.4 Mini
GPT-5.4 Mini leads by 1.0
Metric-level scores with benchmark source, rank depth, confidence, error, and evaluation date where available.
GPT-5
$3.44/M
Rank #117/162 · confidence 4
GPT-5.4 Mini
winner$1.69/M
Rank #95/162 · confidence 4
GPT-5
$1.25/M
Rank #107/162 · confidence 4
GPT-5.4 Mini
winner$0.750/M
Rank #95/162 · confidence 4
GPT-5
$10.00/M
Rank #120/162 · confidence 4
GPT-5.4 Mini
winner$4.50/M
Rank #102/162 · confidence 4
GPT-5
winner75.6
Rank #2/89 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
69.3
Rank #23/89 · confidence 3 · eval date 2026-03-17
GPT-5
5.7
Rank #30/73 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner10.0
Rank #21/73 · confidence 3 · eval date 2026-03-17
GPT-5
67.3%
Rank #121/188 · confidence 4
GPT-5.4 Mini
winner82.3%
Rank #64/188 · confidence 4
GPT-5
5.4%
Rank #132/187 · confidence 4
GPT-5.4 Mini
winner17.1%
Rank #65/187 · confidence 4
GPT-5
20.1
Rank #22/33 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner48.0
Rank #10/33 · confidence 3 · eval date 2026-03-17
GPT-5
42.9
Rank #46/96 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner49.9
Rank #23/96 · confidence 3 · eval date 2026-03-17
GPT-5
37.8
Rank #41/59 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner56.1
Rank #24/59 · confidence 3 · eval date 2026-03-17
GPT-5
38.8%
Rank #83/185 · confidence 4
GPT-5.4 Mini
winner44.2%
Rank #38/185 · confidence 4
GPT-5
82.1
Rank #31/90 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner89.8
Rank #13/90 · confidence 3 · eval date 2026-03-17
GPT-5
winner40.7
Rank #20/90 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
37.5
Rank #32/90 · confidence 3 · eval date 2026-03-17
GPT-5
34.7
Rank #48/99 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner40.0
Rank #31/99 · confidence 3 · eval date 2026-03-17
GPT-5
85.4
Rank #45/96 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner87.5
Rank #34/96 · confidence 3 · eval date 2026-03-17
GPT-5
26.5
Rank #39/96 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner26.6
Rank #38/96 · confidence 3 · eval date 2026-03-17
GPT-5
winner74.2
Rank #32/57 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
73.3
Rank #33/57 · confidence 3 · eval date 2026-03-17
GPT-5
73.1
Rank #29/86 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner73.3
Rank #26/86 · confidence 3 · eval date 2026-03-17
GPT-5
25.7
Rank #28/55 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner30.2
Rank #21/55 · confidence 3 · eval date 2026-03-17
GPT-5
28.9
Rank #31/50 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner33.5
Rank #24/50 · confidence 3 · eval date 2026-03-17
GPT-5
1079.0
Rank #31/55 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
winner1169.0
Rank #24/55 · confidence 3 · eval date 2026-03-17
GPT-5
winner84.8
Rank #44/84 · confidence 1 · eval date 2025-08-07
GPT-5.4 Mini
83.3
Rank #49/84 · confidence 3 · eval date 2026-03-17
Latest completed audits from shared providers, with four safety and integrity score groups plus report links.
| Provider | GPT-5 | GPT-5.4 Mini |
|---|---|---|
Winner: GPT-5.4 Mini | GPT-5 gpt-5-chat-latest No audit yet | GPT-5.4 Mini gpt-5.4-mini Audit score 97 10010086100 |
Winner: GPT-5.4 Mini | GPT-5 gpt-5 No audit yet | GPT-5.4 Mini gpt-5.4-mini Audit score 93 1008486100 |
Speed aggregates and input/output pricing share each provider row for real API selection and migration cost checks.
| Provider | GPT-5 | GPT-5.4 Mini |
|---|---|---|
20 tests | GPT-5 speed / latency 95 tok/s / 1234ms input / output No data | GPT-5.4 Mini speed / latency N/A / N/A input / output No data |
15 tests | GPT-5 speed / latency N/A / N/A input / output No data | GPT-5.4 Mini speed / latency 108 tok/s / 1950ms input / output No data |
10 tests | GPT-5 speed / latency 38 tok/s / 15844ms input / output No data | GPT-5.4 Mini speed / latency 223 tok/s / 3411ms input / output No data |
5 tests | GPT-5 speed / latency N/A / N/A input / output No data | GPT-5.4 Mini speed / latency 123 tok/s / 8529ms input / output No data |
5 tests | GPT-5 speed / latency 46 tok/s / 6577ms input / output No data | GPT-5.4 Mini speed / latency N/A / N/A input / output No data |
GPT-5 gpt-5 speed / latency No data input / output $0/M/$0/M | GPT-5.4 Mini gpt-5.4-mini speed / latency No data input / output $0.062/M | |
GPT-5 gpt-5-all speed / latency No data input / output $0/request | GPT-5.4 Mini gpt-5.4-mini speed / latency No data input / output $0/request |
Weighted outcome: GPT-5.4 Mini. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
Output
| Features | Text inputImage inputFile inputText outputTool callingStructured outputsJSON modeReasoning | Text inputImage inputFile inputText outputTool callingStructured outputsJSON modeReasoning |
|---|
