Model Library
DatabricksReleased March 27, 2024DBRX
Databricks · Released March 27, 2024
Intelligence score
3
At a glance
- Intelligence
- 3
- Speed
- —
- Latency
- —
- Input price
- Not published
- Output price
- Not published
- Context window
- 33Ktokens
- Parameters
- 132billions
Summary
Comparison summary
DBRX Instruct scores 3 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 7).
Specifications
Technical specifications
| Model name | DBRX |
|---|---|
| Provider | Databricks |
| Release date | March 27, 2024 |
| Intelligence score | 3 |
| Speed | — |
| Latency (TTFT) | — |
| Input price | Not published |
| Output price | Not published |
| Context window | 33,000 tokens |
| Parameters | 132B |
| Reasoning | No |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Databricks Open Model License |
| API availability | — |
| Source | Model Intelligence legacy profile: dbrx |
FAQ · 17
Frequently asked questions
DBRX Instruct was released on March 27, 2024.
DBRX Instruct was created by Databricks.
DBRX Instruct scores 3 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 7).
No, DBRX Instruct is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
DBRX Instruct supports text only input.
DBRX Instruct supports text only output.
No, DBRX Instruct does not support image input. It can only process text.
No, DBRX Instruct is not multimodal. It only supports text only input.
DBRX Instruct has a context window of 33k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, DBRX Instruct is open weights. The model weights are publicly available and can be downloaded for self-hosting.
DBRX Instruct has 132 billion parameters (36 billion active).
DBRX Instruct is a Mixture of Experts (MoE) model with 132 billion total parameters, but only 36 billion active parameters are used during inference.
DBRX Instruct is released under the Databricks Open Model License license. This license allows commercial use.
DBRX Instruct achieves a score of 3 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
DBRX Instruct has a knowledge cutoff of December 2023. The model's training data includes information up to this date.
DBRX Instruct is an open weights model that can be self-hosted.
DBRX Instruct is an open weights model that can be downloaded and self-hosted.