GLM-Z1-9B-0414 is a 9.4-billion-parameter model from the new GLM-4-0414 series, which came as a genuine surprise even to its developers. Despite its relatively small parameter count, it was trained using all the key techniques employed in creating much larger models within this series. This approach has resulted in unexpectedly high levels of accuracy, logical reasoning, and overall performance.
The model underwent comprehensive training, starting with pre-training on vast volumes of high-quality data, and continuing through advanced post-training stages (preference alignment, rejection sampling, and reinforcement learning based on pairwise ranking feedback). These methods enabled the model to better understand what constitutes the most helpful and accurate responses in specific situations.
Special emphasis was placed on developing reasoning capabilities during training, particularly for solving mathematical problems and logical puzzles. These skills make the model effective not only in standard question-answer scenarios but also in handling more complex analytical tasks. Furthermore, thanks to its compact size, GLM-Z1-9B-0414 demonstrates excellent computational efficiency, making it an ideal choice for resource-constrained environments.
| 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 | |||
|---|---|---|---|---|---|---|
32,768.0 |
1 | $0.38 | 14.84 | 3.306 | Launch | |
32,768.0 |
1 | $0.53 | 43.50 | 9.120 | Launch | |
32,768.0 |
1 | $0.83 | 78.14 | 9.900 | Launch | |
32,768.0 tensor |
2 | $0.97 | 73.98 | 5.730 | Launch | |
32,768.0 |
1 | $1.02 | 109.82 | 9.871 | Launch | |
32,768.0 tensor |
2 | $1.56 | 25.128 | Launch | ||
32,768.0 |
1 | $1.59 | 15.634 | Launch | ||
32,768.0 |
1 | $2.37 | 71.57 | 50.768 | Launch | |
32,768.0 |
1 | $3.83 | 107.92 | 49.996 | Launch | |
32,768.0 |
1 | $4.11 | 61.074 | Launch | ||
32,768.0 tensor |
2 | $4.61 | 107.104 | Launch | ||
32,768.0 |
1 | $4.74 | 95.453 | Launch | ||
32,768.0 tensor |
2 | $9.40 | 196.234 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
32,768.0 tensor |
2 | $0.57 | 19.28 | 5.363 | Launch | |
32,768.0 |
1 | $0.83 | 43.15 | 1.772 | Launch | |
32,768.0 tensor |
2 | $0.93 | 16.040 | Launch | ||
32,768.0 |
1 | $1.02 | 49.47 | 1.727 | Launch | |
32,768.0 pipeline |
3 | $1.43 | 33.56 | 4.043 | Launch | |
32,768.0 tensor |
2 | $1.56 | 17.600 | Launch | ||
32,768.0 |
1 | $1.59 | 7.514 | Launch | ||
32,768.0 tensor |
4 | $1.82 | 4.086 | Launch | ||
32,768.0 |
1 | $2.37 | 48.25 | 42.720 | Launch | |
32,768.0 |
1 | $3.83 | 42.716 | Launch | ||
32,768.0 |
1 | $4.11 | 52.954 | Launch | ||
32,768.0 tensor |
2 | $4.61 | 99.576 | Launch | ||
32,768.0 |
1 | $4.74 | 87.333 | Launch | ||
32,768.0 tensor |
2 | $9.40 | 188.706 | Launch | ||
Contact our dedicated neural networks support team at support@immers.cloud or send your request to the sales department at sale@immers.cloud.