Цена входных данных
17$0.435/M
17$0.435/M
18DeepSeek V3.2$0.500/M
18$0.500/M
DeepSeek V3.2 by DeepSeek is available through 191 API providers on LMSpeed. Compare API pricing from $0.0010 to $300.00 per million input tokens across providers. Free API access is offered by 9 providers. Context window: 131,072. In speed benchmarks, the fastest provider reaches 272 tok/s.
DeepSeek V3.2 is an upgraded V3-series MoE model with stronger reasoning, coding, and math performance, widely available through OpenAI-compatible API relays.
Также известна как
Compare DeepSeek V3.2 API pricing across 182 providers. Input prices range from $0.0010 to $300.00 per million input. 10dian-API offers the lowest rate at $0.0010/M. 9 providers offer free API credits or a free tier.
| Провайдер | Вариант модели | Аудит | Входные данные ($/M) | Выходные данные ($/M) | Скорость (т/с) | Первый токен |
|---|---|---|---|---|---|---|
6345ywz API99.8% | FAST/deepseek-v3.2 | — | $0.137-73% | $0.411-74% | 255.5 t/s | 1.92 s |
| deepseek-ai/DeepSeek-V3.2 | — | $0.037-93% | $0.058-96% | 23.6 t/s | 3.30 s | |
| Pro/deepseek-ai/DeepSeek-V3.2 | — | $0.037-93% | $0.058-96% | — | — | |
| deepseek-v3.2 | — | $0.079-84% | $0.230-86% | — | — | |
Thorbase100% | deepseek/deepseek-v3.2 | — | $0.217-57% | $0.326-80% | 56.1 t/s | 1.98 s |
紫脑喵99.9% | deepseek-v3.2 | — | $0.010-98% | $0.010-99% | 53.1 t/s | 3.36 s |
| deepseek-v3.2-exp | — | $0.010-98% | $0.010-99% | — | — | |
乐天图书馆100% | deepseek-v3.2 | — | $0.137-73% | $1.37-14% | 49.1 t/s | 6.90 s |
| deepseek-v3.2-exp | — | $0.137-73% | $0.822-49% | — | — | |
| DeepSeek-V3.2-Fast | — | $0.479-4% | $7.19 | — | — | |
| /maas/deepseek-ai/DeepSeek-V3.2 | — | $2.57 | $2.57 | — | — | |
Privnode99.8% | deepseek-v3.2 | — | $0.274-45% | $0.411-74% | 36.1 t/s | 1.40 s |
简小智API中转站98.6% | deepseek-v3.2-exp | — | $1.00 | $1.50-6% | 27.1 t/s | 0.85 s |
| deepseek-ai/DeepSeek-V3.2 | — | $2.00 | $2.00 | — | — | |
| deepseek-ai/DeepSeek-V3.2-Exp | — | $2.00 | $2.00 | — | — | |
| deepseek-v3.2 | — | $2.00 | $3.00 | — | — | |
哈基米API站100% | deepseek-v3.2 | — | $2.00 | $3.00 | 25.7 t/s | 10.23 s |
素墨API100% | deepseek-v3.2 | — | $0.010-98% | $0.010-99% | 21.7 t/s | 5.99 s |
| deepseek-ai/deepseek-v3.2 | — | $0.010-98% | $0.010-99% | — | — | |
| deepseek-v3.2-speciale | — | $0.010-98% | $0.010-99% | — | — |
Данные о ценах из публичных API провайдеров
Сравните производительность по скорости и задержке у всех API-провайдеров.
| Провайдер | Скорость | Задержка | Тесты |
|---|---|---|---|
6345ywz API PRO/deepseek-v3.2 | 272.47 tok/s | 1.09s | 10 |
FAST/deepseek-v3.2 | 255.47 tok/s | 1.92s | 20 |
deepseek-v3.2 | 95.36 tok/s | 3.21s | 10 |
deepseek/deepseek-v3.2 | 56.13 tok/s | 1.98s | 5 |
DeepSeek-V3.2 | 55.34 tok/s | 6.07s | 5 |
deepseek-v3.2 | 53.05 tok/s | 3.36s | 5 |
deepseek/deepseek-v3.2 | 51.09 tok/s | 6.43s | 5 |
deepseek-v3.2 | 49.12 tok/s | 6.90s | 5 |
deepseek-v3.2 | 45.59 tok/s | 6.60s | 5 |
deepseek-v3.2 | 43.44 tok/s | 7.47s | 5 |
deepseek-v3.2 | 42.97 tok/s | 0.96s | 17 |
deepseek-v3.2-exp | 41.00 tok/s | 3.78s | 5 |
deepseek-v3.2-exp | 39.89 tok/s | 0.58s | 5 |
deepseek-v3.2 | 36.61 tok/s | 1.52s | 15 |
deepseek-v3.2 | 36.08 tok/s | 1.40s | 5 |
DeepSeek-V3.2 | 35.11 tok/s | 6.18s | 5 |
Pro/deepseek-ai/DeepSeek-V3.2 | 34.36 tok/s | 22.51s | 40 |
deepseek-v3.2 | 29.74 tok/s | 4.15s | 10 |
deepseek-ai/DeepSeek-V3.2 | 29.48 tok/s | 1.88s | 5 |
deepseek-v3-2-251201 | 29.28 tok/s | 1.94s | 35 |
Показано 1–20 из 44 провайдеров
kimi-k2-5
Moonshot Kimi K2.5 is an open-weight multimodal agent model with native vision and text input, strong coding performance, and a 256K context window.
minimax-m2-5
MiniMax M2.5 is MiniMax's flagship text model for coding and agents, with SOTA-level programming and agentic performance, improved token efficiency, and fast high-TPS API deployment.
gpt-5-4
OpenAI GPT-5.4 extends the GPT-5 family with stronger instruction following, deeper tool use, and improved performance on coding, math, and long-document analysis.
glm-5-1
Zhipu GLM-5.1 is a next-generation GLM model aimed at frontier reasoning, coding, and bilingual agent applications.
deepseek-v4-flash
DeepSeek V4 Flash is a fast, cost-efficient language model in the DeepSeek V4 family, optimized for low-latency chat, coding assistance, and high-throughput API workloads while retaining strong reasoning quality.
glm-5
Zhipu GLM-5 is Zhipu flagship GLM series model with enhanced reasoning, agent capabilities, and strong performance on Chinese enterprise and coding scenarios.
Рейтинги основаны на тестах, предоставленных сообществом, и периодических зондах работоспособности. Носит рекомендательный характер, не является официальными данными.Artificial Analysis
