←
ScitiX253 posts
ScitiX avatar✓

ScitiX ✓

@scitix

Shaping the Future with Intelligent Computing!

◷ Joined September 2024▧ 164 Photos and videos
ScitiX avatar
ScitiX@scitix·Jul 7

Boson AI's flagship speech models are now live on ScitiX Model Inference. bosonai/tts for expressive, conversational text-to-speech, and bosonai/asr for accurate multilingual transcription — a complete voice loop behind one OpenAI-compatible API. Listening and speaking, on infrastructure built to be simple, fast, and secure.

1:05
ScitiX avatar
ScitiX@scitix·Nov 5

Bring Kimi K3.1 Pro into your applications with ScitiX’s OpenAI-compatible inference API. Build with a 1M-token context window for coding, tool use, and long-running agent workflows—through a familiar API interface. Explore Kimi K3.1 Pro on ScitiX: scitix.ai/inference

ScitiX avatar
ScitiX@scitix·Nov 5

Kimi K3.1 Pro is now live on ScitiX. A 3T-class model with a 1M-token context window, built for coding, agentic workflows, 3D creation, and complex research. Try it on ScitiX: scitix.ai/inference

Post attachment
ScitiX avatar
ScitiX@scitix·Oct 19

DeepSeek-V4.1-Flash-1017 is now available on ScitiX. PD-separated serving for stable latency under concurrent load. One endpoint alongside GLM, Qwen, Kimi, MiniMax, and 20+ other models — your architecture stays the same when model requirements change. Zero data retention. 99.9% uptime. T+3 availability. scitix.ai/inference

Post attachmentPost attachmentPost attachment
ScitiX avatar
ScitiX@scitix·Oct 15

Live cluster monitoring, updated in real time. Active jobs, GPU utilization, token throughput, node-level breakdowns, job logs, inference throughput curves—all moving as your workload runs. Not a snapshot. Not a 30-second refresh. Live. Four commands from install to this dashboard. pip install scitix-cli and you're in. → scitix.ai

0:09
ScitiX avatar
ScitiX@scitix·Oct 14

Your AI infra expert, inside Slack, Teams, and your API. Check GPU utilization. Pull price quotes. Set alerts. Trigger autoscale. Troubleshoot a failing job. Manage billing. All without leaving your workspace—natural language or code, whichever fits the moment. No tickets. No portal. Just ask. Connect your workspace: scitix.ai

0:11
ScitiX avatar
ScitiX@scitix·Oct 12

Different tasks need different models. ScitiX Model Advisor shows you why. Complex Reasoning → GLM-5 (744B MoE, long-horizon agentic tasks) Low-Latency RAG → gpt-oss-20b (18ms TTFT, fastest open-weight option) Code, Batch, Vision, Agents & more—describe your use case, get a recommendation with reasoning attached. scitix.ai/inference

0:14
ScitiX avatar
ScitiX@scitix·Oct 3

In MLPerf Inference v6.1, ScitiX recorded 59,668 tok/s Server and 60,410 tok/s Offline for DeepSeek-R1 on 8×SZ002-SXM-180GB. Same hardware, same model—but two different optimization problems. Offline maximizes steady-state throughput with no latency constraint. Server has to protect TTFT and TPOT p99 under continuous request arrival. The scheduler parameters that maximize one can break the other. We tuned them separately. If you're running inference at this scale, that distinction matters more than the headline number.

Post attachment
ScitiX avatar
ScitiX@scitix·Sep 22

Weight movement isn’t solved for trillion-scale models. FlashBoot turns layout, transfer, and replication into one data path—at the bandwidth ceiling. NVL72 · DeepSeek-V4-Pro · TP=4 Load: 20.1s → 0.40s (50× vs InstantTensor) 8 replicas: 87s → 0.32s (270×+ vs SGLang R-Fork)

Post attachmentPost attachment
ScitiX avatar
ScitiX@scitix·Sep 21

FlashBoot co-designs layout, transfer, and replication as one data path—and drives each to hardware limits. Not NVL72-specific: covers C2C/PCIe and NVLink/RDMA, validated on GLM-5.2, DeepSeek-V4, and Kimi-K3. Trillion-parameter elasticity starts here.

ScitiX avatar
ScitiX@scitix·Sep 17

FlashLoad pipelines host→GPU transfer with zero-copy placement, built on FabricArena's single-IO layout. On NVL72: single-node loading 20.1s → 0.40s, up to 50× faster.

Post attachmentPost attachment
ScitiX avatar
ScitiX@scitix·Sep 17

FlashBoot: At measured C2C bandwidth, 217 GB of DeepSeek-V4-Pro weights should transfer in ~1.2s. Baseline loaders took 45–67s—using 3% of available bandwidth.

Post attachment
ScitiX avatar
ScitiX@scitix·Sep 16

Behind these results: a corrected evaluation path, joint tuning of speculative decoding, memory and scheduling, and full-load validation. Our goal was not a short-test peak, but sustainable serving performance within quality and latency constraints. #ModelInference #AIInfra

ScitiX avatar
ScitiX@scitix·Sep 16

This was not unconstrained peak throughput. The Server run remained within the applicable latency limits, with p99 TTFT of 1.719s and p99 TPOT of 70.43ms. The same system also reached 60,410.90 output tokens/s in DeepSeek-R1 Offline (Closed), 3.12% above v6.0.

ScitiX avatar
ScitiX@scitix·Sep 16

LPerf Inference v6.1 results are live. @mlcommons ScitiX delivered 59,668.30 output tokens/s for DeepSeek-R1 Server (Open) on one 8× SZ002-SXM-180GB system—the record-breaking result in its category and 15.43% above the best v6.0 comparison result. mlcommons.org/2026/09/mlperf…