iluvatar-zhikai-mr-v50
Source details
Source revision: 171d06aeae6684ad50dacaf79fffad8d328b394cda86e69032ba0a411362c1e9
Iluvatar CoreX · Data center · Accelerator card
Exact Zhikai 50 MR-V50 reference specifications and AI deployment guidance, separated from FusionPointAI commercial inventory.
Technical diagrams
| Field | Value |
|---|---|
| Manufacturer | Iluvatar CoreX |
| Exact SKU | Zhikai 50 MR-V50 |
| Product family | Iluvatar CoreX Zhikai 50 MR-V50 |
| Architecture | Iluvatar CoreX · Second generation · GPGPU |
| Generation | Zhikai Inference |
| Entity type | Accelerator card |
| Lifecycle | Recent generation |
| Launch date | Not listed |
| VRAM | 16 GB |
| Memory technology | HBM2e |
| Memory bandwidth | Not listed |
| Power | 75 W |
| Interface | PCIe Gen4 x16 |
| Form factor | Half height · Half length · Single slot · PCIe · Accelerator |
| Interconnect | Not listed |
| MIG / partitioning | Not supported |
| Supported numeric formats | FP32, FP16, INT8 |
| Software stack | iluvatar-corex-software-stack, deepspark |
Best suited to smaller checkpoints, compressed models or distributed configurations sized by the recommendation engine.
Designed for sustained accelerator workloads; training suitability depends on interconnect, precision and cluster topology.
FP16 acceleration supports common generative and vector workloads within the listed VRAM envelope.
75 W published power requires chassis, airflow and power delivery for this exact form factor.
No dedicated accelerator interconnect is asserted from the available manufacturer evidence.
iluvatar-corex-software-stack, deepspark
Source revision: 171d06aeae6684ad50dacaf79fffad8d328b394cda86e69032ba0a411362c1e9
This is not a product photograph.
https://www.iluvatar.com/productDetails?fullCode=cpjs-yj-tlxltt-zk100
Last verified: 2026-08-27