slime v0.2.2 expands Int4-QAT and R3 support
Published on January 18, 2026, the slime v0.2.2 version announcement highlights memory and performance improvements, Int4-QAT training, and R3 support with DeepEP and MTP.
Source: slime v0.2.2 adiciona suporte a Int4-QAT e R3 (github.com). Text prepared with AI from this source.
What happened and what to do
The slime v0.2.2 version announcement, published on January 18, 2026, lists memory and performance improvements, Int4-QAT training, and full R3 support with DeepEP and MTP. The version also updates to SGLang v0.5.7 and the Megatron development branch. Engineers using slime can evaluate new training features, rollout support, and performance changes.
A company that trains or serves models can assess these features in a controlled environment, measure memory, performance, and compatibility with its training and rollout workflows, and track changes by version. Wendelmaques can help structure this assessment and, where appropriate, implement monitoring, automation, and components on the company’s own infrastructure.
How the consultancy can help
Wendelmaques can diagnose the opportunity and infrastructure requirements, define a scoped assessment, and implement and operate the necessary components, such as monitoring pipelines and training or rollout automation.
Next step
Send a short description of your case to receive a scoped proposal for diagnosis, implementation, and operation.
Let's assess your project
I reply with scope and a proposal.