BLOOM Models by BigScience
Last refreshed 2026-10-09. Next refresh: weekly.
Details
About
BLOOM is a multilingual large language model from the BigScience collaboration: a 176-billion-parameter autoregressive Transformer trained to continue text from a prompt. It was trained on the French government’s Jean Zay public supercomputer using 1.6 TB of pre-processed text covering 46 natural languages and 13 programming languages, with a 2,048-token sequence length. BigScience released it in July 2022, and smaller BLOOM checkpoints with 560M, 1.1B, 1.7B, 3B, and 7.1B parameters are also available.
Decision facts
- Best fit
- coding
- Capability starting point
- BLOOM 176B with 2k context
- Lowest tracked input
- Not tracked
- Closest related family
- MT0
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs 2k context, 176B parameters, and open source.
Use when the workload needs 2k context, 7.1B parameters, and open source.
Use when the workload needs 2k context, 3B parameters, and open source.
Use when the workload needs 2k context, 1.7B parameters, and open source.
Use when the workload needs 2k context, 1.1B parameters, and open source.
Use when the workload needs 2k context, 560M parameters, and open source.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| BLOOM 176B | Use when the workload needs 2k context, 176B parameters, and open source. | 2022-07 | 2k context176B parametersopen source | Current |
| BLOOM 7B1 | Use when the workload needs 2k context, 7.1B parameters, and open source. | 2022-07 | 2k context7.1B parametersopen source | Current |
| BLOOM 3B | Use when the workload needs 2k context, 3B parameters, and open source. | 2022-07 | 2k context3B parametersopen source | Current |
| BLOOM 1B7 | Use when the workload needs 2k context, 1.7B parameters, and open source. | 2022-07 | 2k context1.7B parametersopen source | Current |
| BLOOM 1B1 | Use when the workload needs 2k context, 1.1B parameters, and open source. | 2022-07 | 2k context1.1B parametersopen source | Current |
| BLOOM 560M | Use when the workload needs 2k context, 560M parameters, and open source. | 2022-07 | 2k context560M parametersopen source | Current |
Release Timeline
1 release groupSpecifications(6 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| BLOOM 176B | 2022-07 | 2k | 176B |
| BLOOM 7B1 | 2022-07 | 2k | 7.1B |
| BLOOM 3B | 2022-07 | 2k | 3B |
| BLOOM 1B7 | 2022-07 | 2k | 1.7B |
| BLOOM 1B1 | 2022-07 | 2k | 1.1B |
| BLOOM 560M | 2022-07 | 2k | 560M |
