Sarvam (reported by MarkTechPost) · India-hosted Inference Platform
Sarvam Token Factory
Sarvam Token Factory is a managed inference platform from Sarvam that serves its own Sarvam 105B model and several open-weight models, including GLM 5.3, DeepSeek V4.1 Flash, Kimi K3.1 Pro and Gemma 4 31B, from GPU clusters in India. Developers reach every model through one OpenAI-compatible API, and the stack uses a cache-aware router, separate prefill and decode GPU pools, and a shared Mooncake-based cache. Sarvam reports gains such as 1.9x per-GPU throughput over InferenceX on Kimi K3.1 Pro, but all benchmark figures are self-reported and Together AI is faster on one DeepSeek test.
Open the source ↗Similar Developer Tools releases
- Managed NVIDIA Dynamo on Amazon SageMaker HyperPod Inference
AWS Machine Learning Blog (Amazon Web Services) · Managed Dynamo inference · Week 48, 2026
- inference-optimisation
sybil-solutions · Inference Optimisation Skill · Week 48, 2026
- SageMaker AI KV cache aware routing
Amazon Web Services (AWS) · KV cache routing · Week 48, 2026
- Together Compiler
Together AI · Agent-driven GPU compiler · Week 48, 2026
- cuBLAS Ozaki-II and 2M FP64 Algorithms
NVIDIA · FP64 GEMM Emulation · Week 47, 2026
- TPU Sync
Google · KV Cache Transfer · Week 47, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.