gpt-5-2
Developer: OpenAI
Also known as
GPT-5.2 by OpenAI is available through 223 API providers on LMSpeed. Compare API pricing from $0.010 to $390098560.00 per million input tokens across providers. Free API access is offered by 2 providers. In speed benchmarks, the fastest provider reaches 94 tok/s.
Compare GPT-5.2 API pricing across 221 providers. Input prices range from $0.010 to $390098560.00 per million input. 素墨API offers the lowest rate at $0.010/M.
| Provider | Model Variant | Input ($/M) | Output ($/M) | Speed (t/s) |
|---|---|---|---|---|
 ZetaTechs API | gpt-5.2-chat-free | Free | Free | — |
| 我不是AI神 | gpt-5.2 | Free | Free | — |
| 素墨API | gpt-5.2 | $0.010 | $0.010 | — |
| 10dian-API | gpt-5.2 | $0.010 | $0.010 | — |
 哈基米API | gpt-5.2 | $0.020 | $0.020 | — |
| CookingAI | gpt-5.2 | $0.030 | $0.030 | — |
| 紫脑喵 | gpt-5.2 | $0.030 | $0.030 | — |
 柏拉图AI | gpt-5.2-all | $0.050 | $0.050 | — |
 毫秒API | gpt-5.2-all | $0.050 | $0.050 | — |
| gpt-5.2-all | $0.050 | $0.050 | — | |
| CookingAI | gpt-5.2-high | $0.050 | $0.050 | — |
| 91VIP API | gpt-5.2 | $0.050 | $0.050 | — |
| 紫脑喵 | gpt-5.2-high | $0.050 | $0.050 | — |
 星见雅 API | gpt-5.2 | $0.100 | $0.800 | 52.3 t/s |
 GLM BigModel Relay | gpt-5.2 | $0.175 | $1.40 | — |
| GLM BigModel Relay | gpt-5.2-high | $0.175 | $1.40 | — |
 VVCode | gpt-5.2 | $0.175 | $1.40 | — |
| Yunchu API | gpt-5.2 | $0.240 | $1.92 | — |
 OpenRouter Fans | gpt-5.2 | $0.250 | $2.00 | — |
 MonkingAI | gpt-5.2 | $0.280 | $2.24 | — |
Pricing data from provider public APIs
GPT-5.2 is free to use through 2 providers with no per-token charges. These providers offer free API credits or a free tier:
| Provider | Speed (t/s) |
|---|---|
| ZetaTechs API | — |
| 我不是AI神 | — |
Compare speed and latency performance across all API providers.
Showing 1-22 of 22 providers
| Provider | Speed | Latency | Tests |
|---|---|---|---|
 PackyAPIgpt-5.2 | 93.71 tok/s | 11.49s | 10 |
Unified AI Gateway gpt-5.2-chat | 71.48 tok/s | 2.89s | 5 |
gpt-5.2 | 64.56 tok/s | 4.77s | 5 |
 gpt-5.2 | 64.23 tok/s | 3.55s | 5 |
 gpt-5.2 | 64.21 tok/s | 5.26s | 5 |
 gpt-5.2 | 64.17 tok/s | 2.95s | 5 |
 gpt-5.2 | 60.05 tok/s | 3.33s | 5 |
 gpt-5.2 | 58.71 tok/s | 1.63s | 10 |
gpt-5.2 | 57.02 tok/s | 4.60s | 5 |
 gpt-5.2 | 56.63 tok/s | 1.90s | 5 |
gpt-5.2 | 53.95 tok/s | 1.45s | 5 |
 gpt-5.2 | 53.46 tok/s | 1.88s | 10 |
 gpt-5.2 | 52.28 tok/s | 2.67s | 15 |
gpt-5.2-chat | 51.41 tok/s | 2.14s | 5 |
gpt-5.2 | 50.67 tok/s | 2.72s | 5 |
gpt-5.2 | 48.12 tok/s | 2.93s | 10 |
gpt-5.2 | 48.10 tok/s | 1.54s | 5 |
 gpt-5.2 | 46.23 tok/s | 8.11s | 5 |
GPT-5.2 | 45.37 tok/s | 1.95s | 5 |
gpt-5.2 | 44.55 tok/s | 1.72s | 5 |
gpt-5.2 | 39.67 tok/s | 3.20s | 5 |
 gpt-5.2-2025-12-11 | 32.37 tok/s | 1.97s | 5 |
Latest benchmark results measuring API response speed and first-token latency.
| Time | Model | Speed | Latency |
|---|---|---|---|
| 03/28/2026, 10:52 | gpt-5.2 | 55.10 tok/s | 1.39s |
| 03/28/2026, 10:52 | gpt-5.2 | 49.50 tok/s | 1.27s |
| 03/28/2026, 10:52 | gpt-5.2 | 54.41 tok/s | 1.43s |
| 03/28/2026, 10:52 | gpt-5.2 | 53.78 tok/s | 1.30s |
| 03/28/2026, 10:52 | gpt-5.2 | 56.98 tok/s | 1.85s |
| 03/26/2026, 08:57 | gpt-5.2 | 55.20 tok/s | 5.73s |
| 03/26/2026, 08:57 | gpt-5.2 | 47.74 tok/s | 2.12s |
| 03/26/2026, 08:57 | gpt-5.2 | 105.29 tok/s | 10.75s |
| 03/26/2026, 08:57 | gpt-5.2 | 57.99 tok/s | 3.93s |
| 03/26/2026, 08:57 | gpt-5.2 | 56.58 tok/s | 1.31s |
| 03/26/2026, 08:56 | gpt-5.2 | 52.82 tok/s | 7.60s |
| 03/26/2026, 08:56 | gpt-5.2 | 46.05 tok/s | 6.58s |
| 03/26/2026, 08:56 | gpt-5.2 | 30.33 tok/s | 19.98s |
| 03/26/2026, 08:56 | gpt-5.2 | 45.19 tok/s | 3.14s |
| 03/26/2026, 08:56 | gpt-5.2 | 56.76 tok/s | 3.23s |
| 03/24/2026, 06:23 | gpt-5.2 | 54.46 tok/s | 1.08s |
| 03/24/2026, 06:23 | gpt-5.2 | 49.65 tok/s | 2.43s |
| 03/24/2026, 06:23 | gpt-5.2 | 53.66 tok/s | 1.36s |
| 03/24/2026, 06:23 | gpt-5.2 | 43.97 tok/s | 2.91s |
| 03/24/2026, 06:23 | gpt-5.2 | 55.95 tok/s | 1.07s |
