Download release.json from DevelopingDad/GLM-5.3-Flash-FP4-Local-DGX-Spark-Optimized: direct link, hf CLI and curl.
- Browser
- Download file 2.25 kB
-
https://huggingface.co/DevelopingDad/GLM-5.3-Flash-FP4-Local-DGX-Spark-Optimized/resolve/main/release.json
- Command line
-
hf download hf://DevelopingDad/GLM-5.3-Flash-FP4-Local-DGX-Spark-Optimized/release.json
-
curl -L -o release.json https://huggingface.co/DevelopingDad/GLM-5.3-Flash-FP4-Local-DGX-Spark-Optimized/resolve/main/release.json
2.25 kB
| { | |
| "release": "qualified-20261024", | |
| "base_model": "nvidia/GLM-5.3-Flash-FP4-Local", | |
| "model_revision": "76348ddb2be2e35e39ecfc00e75dc07703f4f3d5", | |
| "target_weight_changes": false, | |
| "target_quantization_changes": false, | |
| "model_configuration_change": "Explicit terminal EOS ID 154828 added to the upstream generation policy.", | |
| "image": "eugr/spark-vllm-b12x@sha256:03a956544e9f9b4b7ba91690469fbe7848bdbef29a7d2e0f5788368467bb3595", | |
| "image_id": "sha256:a7f76ae9f18c7701c9cde5f72c606393d9a21ca94dd8e53b27589935f64fcbe2", | |
| "vllm_version": "0.1.dev21830+g8244d1c44.d20261015", | |
| "patches": [ | |
| { | |
| "target": "vllm/model_executor/layers/quantization/nvfp4_csf.py", | |
| "patch": "patches/vllm_model_executor_layers_quantization_nvfp4_csf.py.patch", | |
| "original_sha256": "0ae697a86efb41a1149885b88499c878636da54cafd9f2bb5710ea4626062c5c", | |
| "qualified_sha256": "c9035eeafaf35ce6a1cd65eecd05515b295e7ed4543694e0486c228de7ed2d1e" | |
| }, | |
| { | |
| "target": "vllm/v1/spec_decode/dynamic/acceptance_length.py", | |
| "patch": "patches/vllm_v1_spec_decode_dynamic_acceptance_length.py.patch", | |
| "original_sha256": "2e3167f3a16006c7de34d6eda78aae1bc0a3489171b17c41122cd782303792c8", | |
| "qualified_sha256": "60a5917beca3ffcd89e6d3e01cde306d4a31c8aa02a32a4576f43c8ba4d07c9d" | |
| }, | |
| { | |
| "target": "vllm/config/vllm.py", | |
| "patch": "patches/vllm_config_vllm.py.patch", | |
| "original_sha256": "a7e678c5d65d14c64db350808e3d979d3643412b7f9d1733e3434703da6afda2", | |
| "qualified_sha256": "56bb042299667fa4687fc85c76c5270037fd0f55b28cd40d3f20fe125b7762c3" | |
| }, | |
| { | |
| "target": "vllm/v1/worker/gpu/spec_decode/autoregressive/speculator.py", | |
| "patch": "patches/vllm_v1_worker_gpu_spec_decode_autoregressive_speculator.py.patch", | |
| "original_sha256": "59eed0165a962c05545403e6616e3bc5c8b036e8b87feef93cd9e82748bbf887", | |
| "qualified_sha256": "d60cd4fbc5fd63f937815c8c0088906484e18d078f46bc8186f01736764a9847" | |
| } | |
| ], | |
| "generation_config_sha256": "20473c0f9abc10ed7137f1ba6522288753c986e9b33acb18e774ae411b6dcfb5", | |
| "context": 262144, | |
| "kv_bytes_per_rank": 8589934592, | |
| "reported_logical_kv_tokens": 808020, | |
| "max_sequences": 4, | |
| "speculation": "Native MTP; DFlash is not part of this release." | |
| } | |