Phi-4 is a modern open-source language model with 14 billion parameters. Although the architectural changes compared to previous versions are minimal, the model demonstrates significant progress in tasks requiring logical thinking and analytical skills, made possible by an innovative training approach. Unlike traditional language model training methods, Phi-4 focuses not on the quantity of data but on its quality. The training of Phi-4 utilized diverse sources, including synthetic data specifically designed to develop reasoning skills, filtered documents from public sources, as well as acquired academic books and question-answer knowledge bases. This allows the model to achieve high performance even with a relatively small size.
Phi-4 works exclusively with textual data. Its context window is relatively small at 16K tokens, but it supports more than 50 languages, including Russian.
Overall, Phi-4 is a versatile lightweight model, but according to its developers, it is particularly effective in environments with limited memory and computational resources, as well as for tasks requiring instant response.
| Model Name | Context | Type | GPU | Status | Link |
|---|
There are no public endpoints for this model yet.
An endpoint is a server with pre-installed software that automatically downloads neural network weights and launches an OpenAI-compatible API to handle incoming requests.
A private endpoint is created within the user's project, belongs exclusively to the user, and is billed per server configuration in accordance with the infrastructure pricing.
We recommend creating private endpoints when you need to:
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
16,384.0 |
1 | $0.53 | 32.82 | 2.752 | Launch | |
16,384.0 tensor |
2 | $0.57 | 9.15 | 4.456 | Launch | |
16,384.0 |
1 | $0.83 | 50.09 | 3.064 | Launch | |
16,384.0 |
1 | $1.02 | 69.85 | 3.052 | Launch | |
16,384.0 pipeline |
3 | $1.43 | 2.627 | Launch | ||
16,384.0 tensor |
2 | $1.56 | 9.066 | Launch | ||
16,384.0 |
1 | $1.59 | 83.03 | 5.361 | Launch | |
16,384.0 |
1 | $2.37 | 53.63 | 19.459 | Launch | |
16,384.0 |
1 | $3.83 | 63.45 | 19.272 | Launch | |
16,384.0 |
1 | $4.11 | 91.22 | 23.358 | Launch | |
16,384.0 tensor |
2 | $4.61 | 41.856 | Launch | ||
16,384.0 |
1 | $4.74 | 37.109 | Launch | ||
16,384.0 tensor |
2 | $9.40 | 77.508 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
16,384.0 |
1 | $0.53 | 34.30 | 1.350 | Launch | |
16,384.0 tensor |
2 | $0.57 | 22.23 | 3.131 | Launch | |
16,384.0 |
1 | $0.83 | 50.17 | 1.854 | Launch | |
16,384.0 |
1 | $1.02 | 43.48 | 1.574 | Launch | |
16,384.0 pipeline |
3 | $1.43 | 43.33 | 2.527 | Launch | |
16,384.0 tensor |
2 | $1.56 | 8.407 | Launch | ||
16,384.0 |
1 | $1.59 | 4.148 | Launch | ||
16,384.0 |
1 | $2.37 | 80.23 | 18.042 | Launch | |
16,384.0 |
1 | $3.83 | 96.69 | 17.941 | Launch | |
16,384.0 |
1 | $4.11 | 143.66 | 22.033 | Launch | |
16,384.0 tensor |
2 | $4.61 | 41.197 | Launch | ||
16,384.0 |
1 | $4.74 | 36.076 | Launch | ||
16,384.0 tensor |
2 | $9.40 | 76.849 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
16,384.0 tensor |
2 | $0.93 | 29.00 | 3.290 | Launch | |
16,384.0 pipeline |
3 | $1.06 | 5.77 | 2.053 | Launch | |
16,384.0 tensor |
2 | $1.56 | 46.56 | 4.138 | Launch | |
16,384.0 tensor |
2 | $1.56 | 4.138 | Launch | ||
16,384.0 tensor |
2 | $1.92 | 55.00 | 4.057 | Launch | |
16,384.0 |
1 | $2.37 | 46.21 | 13.984 | Launch | |
16,384.0 tensor |
2 | $2.93 | 73.39 | 8.468 | Launch | |
16,384.0 |
1 | $3.83 | 53.09 | 13.797 | Launch | |
16,384.0 |
1 | $4.11 | 87.21 | 17.137 | Launch | |
16,384.0 tensor |
2 | $4.61 | 36.868 | Launch | ||
16,384.0 |
1 | $4.74 | 31.876 | Launch | ||
16,384.0 tensor |
2 | $9.40 | 72.520 | Launch | ||
Contact our dedicated neural networks support team at support@immers.cloud or send your request to the sales department at sale@immers.cloud.