Скорость вывода
10040.5 tok/s
10139.3 tok/s
102MiMo-V2.5-Pro38.7 tok/s
10337.1 tok/s
MiMo-V2.5-Pro is available through 61 API providers on LMSpeed. Compare API pricing from $0.0000 to $547.50 per million input tokens across providers. Free API access is offered by 1 provider. Context window: 1,048,576. In speed benchmarks, the fastest provider reaches 64 tok/s.
Xiaomi MiMo-V2.5-Pro is a large open-source language model in the MiMo series, offering advanced reasoning and general-purpose capabilities.
Также известна как
Compare MiMo-V2.5-Pro API pricing across 60 providers. Input prices range from $0.0000 to $547.50 per million input. VSLLM offers the lowest rate at $0.0000/M. 1 provider offers free API credits or a free tier.
| Провайдер | Вариант модели | Аудит | Входные данные ($/M) | Выходные данные ($/M) | Скорость (т/с) | Первый токен |
|---|---|---|---|---|---|---|
Tokeness99.9% | mimo-v2.5-pro | — | $0.325-25% | $0.975 | 63.7 t/s+65% | 3.18 s |
PICO AI47.8% | mimo-v2.5-pro | — | $7.35 | $22.05 | 63.1 t/s+63% | 2.76 s |
小水管 API100% | mimo-v2.5-pro | — | Бесплатно | Бесплатно | 62.7 t/s+62% | 2.55 s |
PackyAPI100% | mimo-v2.5-pro | — | $0.429-1% | $0.857-1% | — | — |
柏拉图AI100% | mimo-v2.5-pro | — | $0.959 | $2.88 | — | — |
CatClaw API100% | mimo-v2.5-pro | — | $0.0040-99% | $0.0040-100% | — | — |
星见雅 API100% | mimo-v2.5-pro | — | $0.100-77% | $0.100-89% | — | — |
天絮 API100% | mimo-v2.5-pro | — | $7.14 | $21.43 | — | — |
Cuz AI100% | mimo-v2.5-pro | 6910063100 | $1.00 | $3.00 | — | — |
速创API99.9% | mimo-v2.5-pro | — | $0.454 | $0.910 | — | — |
6345ywz API99.8% | mimo-v2.5-pro | — | $0.137-69% | $0.411-53% | — | — |
VSLLM99.7% | mimo-v2.5-pro | — | $0.0000-100% | $0.0000-100% | — | — |
ChooseC API99.7% | mimo-v2.5-pro | — | $3.00 | $6.00 | — | — |
RenRen API99.5% | mimo-v2.5-pro | — | $75.00 | $75.00 | — | — |
初叶🍂Furry API98.1% | mimo-v2.5-pro | — | $0.050-89% | $0.050-94% | — | — |
Venlacy100% | mimo-v2.5-pro | — | $0.0010-100% | $0.0010-100% | — | — |
OpenRouter Fans100% | mimo-v2.5-pro | — | $0.050-89% | $0.050-94% | — | — |
| mimo-v2.5-pro | — | $3.18 | $6.35 | — | — | |
17NAS API100% | mimo-v2.5-pro | — | $0.077-82% | $0.230-74% | — | — |
| mimo-v2.5-pro | — | $1.00 | $3.00 | — | — |
Данные о ценах из публичных API провайдеров
Сравните производительность по скорости и задержке у всех API-провайдеров.
| Провайдер | Скорость | Задержка | Тесты |
|---|---|---|---|
Tokeness mimo-v2.5-pro | 63.70 tok/s | 3.18s | 5 |
mimo-v2.5-pro | 63.11 tok/s | 2.76s | 5 |
mimo-v2.5-pro | 62.73 tok/s | 2.55s | 5 |
XiaomiMiMo/MiMo-V2.5-Pro | 60.90 tok/s | 6.13s | 5 |
mimo-v2.5-pro | 59.81 tok/s | 1.71s | 5 |
mimo-v2.5-pro | 45.03 tok/s | 7.94s | 50 |
mimo-v2.5-pro | 44.30 tok/s | 5.05s | 115 |
mimo-v2.5-pro | 43.95 tok/s | 2.77s | 5 |
Показано 1–8 из 8 провайдеров
mimo-v2-5
Xiaomi MiMo-V2.5 is a native omnimodal sparse MoE model (310B total, 15B active) with unified text, image, video, and audio understanding, built on the MiMo-V2-Flash backbone with dedicated vision and audio encoders. It supports up to 1M tokens of context, strong agentic workflows, and open weights on Hugging Face.
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.
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.
deepseek-v4-pro
DeepSeek V4 Pro is the professional-tier DeepSeek V4 model, targeting frontier reasoning, coding, and agent workflows with maximum capability.
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.
glm-5-1
Zhipu GLM-5.1 is a next-generation GLM model aimed at frontier reasoning, coding, and bilingual agent applications.
Рейтинги основаны на тестах, предоставленных сообществом, и периодических зондах работоспособности. Носит рекомендательный характер, не является официальными данными.Artificial Analysis
