Bespoke Labs
About Bespoke Labs
Bespoke Labs develops training environments, data research and supporting infrastructure for AI agents. Its public materials describe reinforcement-learning environments, policy optimization and research on training data. These resources address how agents are trained and exercised as well as the models themselves.
Focus
- Agent-training environments
- Data curation research
- Policy optimization
Projects and interfaces
Reinforcement-learning environments
training infrastructure
Simulation and code-oriented environments described for training longer agent workflows.
GEPA and OpenThoughts
research projects
Optimization and data projects referenced in the company's published research materials.
Scope and limitations
- Provider statements about training or reliability remain separate from Sigma5C's observations.
Official references
- Bespoke Labs official website. Checked .
Models
36 named catalog records. These are name and distribution records, not a count of distinct model architectures.
- bespoke-minicheck
- bespoke-minicheck:7b
- bespoke-minicheck:7b-fp16
- bespoke-minicheck:7b-q2_K
- bespoke-minicheck:7b-q3_K_L
- bespoke-minicheck:7b-q3_K_M
- bespoke-minicheck:7b-q3_K_S
- bespoke-minicheck:7b-q4_0
- bespoke-minicheck:7b-q4_1
- bespoke-minicheck:7b-q4_K_M
- bespoke-minicheck:7b-q4_K_S
- bespoke-minicheck:7b-q5_0
- bespoke-minicheck:7b-q5_1
- bespoke-minicheck:7b-q5_K_M
- bespoke-minicheck:7b-q5_K_S
- bespoke-minicheck:7b-q6_K
- bespoke-minicheck:7b-q8_0
- bespoke-minicheck:latest
- nimble
- nimble-v3
- nimble-v3:9b
- nimble-v3:9b-bf16
- nimble-v3:9b-mlx-bf16
- nimble-v3:9b-mxfp8
- nimble-v3:9b-nvfp4
- nimble-v3:9b-q4_K_M
- nimble-v3:9b-q8_0
- nimble-v3:latest
- nimble:9b
- nimble:9b-bf16
- nimble:9b-mlx-bf16
- nimble:9b-mxfp8
- nimble:9b-nvfp4
- nimble:9b-q4_K_M
- nimble:9b-q8_0
- nimble:latest