{"name":"GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold","repo":"MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold","url":"https://midorreal.com/project/miaai-lab-glm-5-3-flash-exl3-2x-dgx-sparks-tensorfold","repository":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold","description":"GLM-5.3-Flash EXL3 on 2x DGX Spark with TensorFold","facts":{"license":{"value":"Apache-2.0","source":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold","checked":"22 Nov 2026"},"language":{"value":"Python","source":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold","checked":"22 Nov 2026"},"homepage":{"value":"https://mia-ai.net","source":"https://mia-ai.net","checked":"22 Nov 2026"},"releases":{"value":"0","source":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold/releases","checked":"21 Nov 2026"},"contributors":{"value":"24","source":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold/graphs/contributors","checked":"21 Nov 2026"},"archived":{"value":"no","source":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold","checked":"22 Nov 2026"}},"stars":535.0,"writeup":{"what_it_is":"A recipe for serving the GLM-5.3-Flash model from two NVIDIA DGX Sparks through an OpenAI-compatible API, using the TensorFold runtime in NVIDIA's PyTorch container. It supports 4 concurrent requests, a 1,048,576-token context, image and video input, tool calling, and structured outputs. The README also states that this recipe is superseded by a newer one.","audience":"Users with two NVIDIA DGX Sparks who want to run GLM-5.3-Flash locally","claims":[{"kind":"specific","claim":"Serves the model from two DGX Sparks through an OpenAI-compatible API with 4 concurrent requests","excerpt":"Serve **GLM-5.3-Flash** from two NVIDIA DGX Sparks","status":"not_checked"},{"kind":"specific","claim":"Supports the model's full 1,048,576-token context","excerpt":"the model's full **1,048,576-token context**","status":"not_checked"},{"kind":"specific","claim":"Reports decode of 60.4 tok/s for one prose request","excerpt":"| 1 | 60.4 tok/s | 60.4 tok/s | 170 ms |","status":"not_checked"},{"kind":"specific","claim":"Reports 98.8% on GSM8K with 250 problems and thinking off","excerpt":"GSM8K (250 problems, thinking off) | 98.8%","status":"not_checked"},{"kind":"broad","claim":"Describes its performance as fast, a broad claim","excerpt":"Serve **GLM-5.3-Flash** from two NVIDIA DGX Sparks","status":"too_broad"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-11-05"},"why_now":null,"owner_supplied":null,"score":{"verdict":"sleeper","hype":24,"reality":66,"gap":-42,"momentum":0.3,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T09:00:00+00:00","verdict_source":"rule"},"early_read":null,"changes":[{"date":"2026-09-30T22:38:58+00:00","text":"Repository created","source":"https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold"}],"cite":"GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold is a GitHub project at https://github.com/MiaAI-Lab/GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold written mainly in Python, described there as \"GLM-5.3-Flash EXL3 on 2x DGX Spark with TensorFold\".","attribution":"Data from Mid or Real, https://midorreal.com/project/miaai-lab-glm-5-3-flash-exl3-2x-dgx-sparks-tensorfold, checked 22 November 2026. Attribution with a link is required.","still_real":null}