GLM-5.3-Flash on 4× NVIDIA CMP 170HX with vLLM: up to 482.6 tok/s single-user, up to 943.7 tok/s at 8 users, 262K context, TP4 or PP4, one-command container install
☆77Oct 16, 2026Updated this week
Alternatives and similar repositories for glm53-flash-cmp170hx-recipe
Users that are interested in glm53-flash-cmp170hx-recipe are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Tuning and qualification harness for the NVIDIA CMP 170HX (GA100). Measures, gates and recovers clock/voltage settings, and treats a comp…☆89Sep 28, 2026Updated 2 weeks ago
- Community-tested recipes, diagnostics, and benchmarks for CMP 170HX AI workloads☆84Oct 10, 2026Updated last week
- DeepSeek-V4-Flash-0731 on 4x CMP 170HX (sm_80): 98 tok/s decode, ~5300 tok/s prefill. DSpark speculative decoding under pipeline parallel…☆118Aug 13, 2026Updated 2 months ago
- CPU-GPU hybrid inference for DeepSeek-V4 on NVIDIA SM80+ (A100/RTX 4090 etc.), forked from yhfgyyf/vllm-deepseek-v4-sm89.☆60Sep 16, 2026Updated last month
- Validated GLM-5.3 Flash recipe for 2x NVIDIA RTX PRO 6000 Blackwell 96GB: 262K context, EXL3/TR3, adaptive MTP, tools, and vision.☆58Sep 20, 2026Updated 3 weeks ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- A tool to unlobotomize your NVIDIA card!☆773Oct 8, 2026Updated last week
- OREN: Octree Residual Network for Real-Time Euclidean Signed Distance Mapping☆41Jul 8, 2026Updated 3 months ago
- Switchless RoCE collectives, NCCL profiles and fabric tooling for DGX Spark inference recipes☆37Updated this week
- Serve the official Ling-3.0-flash INT4 on one NVIDIA DGX Spark, 38.7 tok/s, faster than the community GGUF. Recipe, watchdog, benchmarks.☆26Aug 29, 2026Updated last month
- DeepSeek V4.1 Flash on 4x RTX PRO 6000 Blackwell: native weights, DSpark, vision, NVMe or locked RAM Engram offload.☆73Updated this week
- Serve Ornith-1.5-35B-A3B (NVFP4 + in-checkpoint MTP, 256k ctx) on one DGX Spark — vLLM b12x recipe plus CUDA-graph fixes.☆32Aug 19, 2026Updated last month
- Audio recorder and transcriber. Store audio files and text to website.☆11Jul 24, 2025Updated last year
- GLM-5.3 (full) on 3-6 DGX Sparks (GB10): a from-scratch CUDA inference engine☆37Updated this week
- Batched, checkpointed, resumable data operations for Laravel — backfills, cleanups, and migrations that survive deploys, crashes, and Ctr…☆35Aug 14, 2026Updated 2 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Unofficial PoC: a BERT-style encoder decision engine behind a typed-decision (noul / choice / score) HTTP API. Not affiliated with TypeSa…☆35Sep 22, 2026Updated 3 weeks ago
- Experimental Drosophila-inspired camera-to-robot bridge. Working mock neural backend, optical flow, IMU feedback and UDP motor control. M…☆229Updated this week
- Two-node DGX Spark/ASUS GX10 DeepSeek V4 Flash DSpark NVFP4 port for vLLM 0.25, with live dashboard and reproducible deployment.☆81Sep 13, 2026Updated last month
- Qwen3.8-27B on one DGX Spark: SGLang + DFlash2 recipe, engine image, benchmarks, data and charts. By @redp314☆37Sep 6, 2026Updated last month
- GLM-5.3 Flash EXL3 for 2-4x DGX Sparks☆736Updated this week
- Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM: 127 tok/s single-user (381 when …☆2,054Updated this week
- An open-source AI hedge fund. A panel of AI analysts and quant models research tickers, each forms a view, and the portfolio layer turns …☆20Jul 27, 2026Updated 2 months ago
- Diablo II: Resurrected on iPhone, iPad & Apple silicon Mac. Play offline with touch or a controller, one-tap Runewords and 100-level Grea…☆165Updated this week
- Your git repos as books on a 3D bookshelf. Browse, search, move, rename, open.☆18Sep 9, 2026Updated last month
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- A Rust library for draping polygons over terrain with gfx☆11Mar 5, 2018Updated 8 years ago
- Cross-platform desktop screen recorder + post-recording editor. Record (with system audio, mic, webcam), then re-frame: backgrounds, layo…☆36Updated this week
- LLM Inference Engine for Metal and CUDA.☆1,375Updated this week
- DNS Checker API with unlimited use, no restrictions applied. Open Source.☆20Nov 26, 2024Updated last year
- An always-on personal agent you own, on the model you choose. A model-agnostic alternative to OpenAI's Dots, built on OpenComputer Server…☆24Sep 29, 2026Updated 2 weeks ago
- Strata fork that runs GLM-5.3-Flash (320B-A18B, 157 GB) on one PC: RTX 5090 + 128 GB RAM + NVMe, ~30 tok/s chat. Fork of Niko1221/Strata …☆32Updated this week
- GLM-5.2-Int4-Int8 on 8x GB10 cluster☆39Jul 30, 2026Updated 2 months ago
- Use OpenCode in WeChat☆23Mar 22, 2026Updated 6 months ago
- Sushi engine runs very high quality quant model in Apple Silicon, using hybrid affine/exl3 with special expert tuning techniques.☆280Updated this week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Metal Cuda Direct Memory Access☆197Updated this week
- ☆81Jun 1, 2026Updated 4 months ago
- MQTT client to control BroadLink RM devices☆11Dec 30, 2020Updated 5 years ago
- DeepSeek 4 Flash local inference engine for Metal☆74Updated this week
- the exit desk for robinhood chain · a sniper, a guard, a follow · reads by default, signs when you arm it☆56Sep 7, 2026Updated last month
- 레트로 게임 한글 패치 제작 전 파이프라인 에이전트 스킬 · Agent skill for end-to-end Korean (Hangul) fan-translation patches for retro games☆91Oct 8, 2026Updated last week
- ☆23Dec 15, 2016Updated 9 years ago