Skip to content

Actions: NVIDIA/Model-Optimizer

Actions

Unit tests

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
unit_tests.yml
2,500+ workflow runs
2,500+ workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Add imatrix calibration for the GGML formats and use it for fused MoE experts Unit tests #13436: Pull request #2735 synchronize by cjluo-nv
In progress chenjiel/iq-imatrix-calibration
chenjiel/iq-imatrix-calibration
In progress
Add a NemotronH omni VL W4A4 PTQ + QAD tutorial for Megatron-Bridge Unit tests #13435: Pull request #2720 synchronize by yueshen2016
In progress yueshen/super-vl-tutorial
yueshen/super-vl-tutorial
In progress
Add GLM-5.3-Flash indexer scorer keyword arguments to vLLM fake-quant serving Unit tests #13434: Pull request #2725 synchronize by sychen52
In progress sychen52:vllm-indexer-scorer-kwargs
sychen52:vllm-indexer-scorer-kwargs
In progress
Fix the FSDP2 export deadlock at world > 1 Unit tests #13433: Pull request #2731 synchronize by cjluo-nv
28m 47s chenjiel/fix-fsdp2-streaming-deadlock
chenjiel/fix-fsdp2-streaming-deadlock
28m 47s
Add a NemotronH omni VL W4A4 PTQ + QAD tutorial for Megatron-Bridge Unit tests #13432: Pull request #2720 synchronize by yueshen2016
35m 59s yueshen/super-vl-tutorial
yueshen/super-vl-tutorial
35m 59s
Add MoE dispatch/combine fake quantization for vLLM's DeepGEMM Mega-MoE Unit tests #13431: Pull request #2767 opened by sychen52
31m 45s sychen52:ep_comms_megamoe
sychen52:ep_comms_megamoe
31m 45s
Make folding exact for GPTQ'd GGML weights, so the vLLM fake-quant ex… Unit tests #13430: Commit 89632ac pushed by cjluo-nv
17m 52s main
main
17m 52s
docs: add AutoQuant KV mixed-precision quantization blog Unit tests #13429: Pull request #2712 synchronize by meenchen
18m 33s docs/autoquant-kv-blog
docs/autoquant-kv-blog
18m 33s
Add a NemotronH omni VL W4A4 PTQ + QAD tutorial for Megatron-Bridge Unit tests #13428: Pull request #2720 synchronize by yueshen2016
29m 12s yueshen/super-vl-tutorial
yueshen/super-vl-tutorial
29m 12s
Add a NemotronH omni VL W4A4 PTQ + QAD tutorial for Megatron-Bridge Unit tests #13427: Pull request #2720 synchronize by yueshen2016
10m 27s yueshen/super-vl-tutorial
yueshen/super-vl-tutorial
10m 27s
fix(onnx): move a negative DequantizeLinear axis when transposing the weight Unit tests #13426: Pull request #2583 synchronize by SID-6921
-1s SID-6921:fix/dq-transpose-negative-axis
SID-6921:fix/dq-transpose-negative-axis
-1s
fix(export): drop reverse renames that invent names the source checkpoint lacks Unit tests #13425: Pull request #2766 opened by juhi10071998
34m 27s bug_6914213
bug_6914213
34m 27s
[2/6] Torch GDN/KDA decode QAT with INT8 recurrent state Unit tests #13424: Pull request #2519 synchronize by kaix-nv
32m 43s kaix/linear-attention-decode-first
kaix/linear-attention-decode-first
32m 43s
fix(export): drop reverse renames that invent names the source checkpoint lacks Unit tests #13423: Pull request #2765 opened by juhi10071998
29m 47s bug_6914213
bug_6914213
29m 47s
Add a NemotronH aggressive-mse NVFP4 PTQ recipe (#2703) Unit tests #13422: Commit 5b01d53 pushed by juhi10071998
18m 39s main
main
18m 39s
Make folding exact for GPTQ'd GGML weights, so the vLLM fake-quant export evaluates them Unit tests #13421: Pull request #2710 synchronize by cjluo-nv
33m 25s chenjiel/ggml-upcast-bf16
chenjiel/ggml-upcast-bf16
33m 25s
fix(onnx): move a negative DequantizeLinear axis when transposing the weight Unit tests #13420: Pull request #2583 synchronize by SID-6921
Action required SID-6921:fix/dq-transpose-negative-axis
SID-6921:fix/dq-transpose-negative-axis
Action required
fix(onnx): move a negative DequantizeLinear axis when transposing the weight Unit tests #13419: Pull request #2583 synchronize by SID-6921
Action required SID-6921:fix/dq-transpose-negative-axis
SID-6921:fix/dq-transpose-negative-axis
Action required
Upload lm_eval results for qwen 3 8B (mmlu + wikitext) Unit tests #13418: Pull request #2763 synchronize by danielkorzekwa
17m 55s dkorzekws/hf_qad_nel_results
dkorzekws/hf_qad_nel_results
17m 55s
Add a pre-push review pass to the agent instructions Unit tests #13417: Pull request #2764 opened by cjluo-nv
12m 15s chenjiel/agents-pre-push-review
chenjiel/agents-pre-push-review
12m 15s
Make folding exact for GPTQ'd GGML weights, so the vLLM fake-quant export evaluates them Unit tests #13416: Pull request #2710 synchronize by cjluo-nv
31m 56s chenjiel/ggml-upcast-bf16
chenjiel/ggml-upcast-bf16
31m 56s
Add a NemotronH aggressive-mse NVFP4 PTQ recipe Unit tests #13415: Pull request #2703 synchronize by juhi10071998
33m 44s nemotron_h_aggressive_recipe
nemotron_h_aggressive_recipe
33m 44s
Add imatrix calibration for the GGML formats and use it for fused MoE experts Unit tests #13414: Pull request #2735 synchronize by cjluo-nv
33m 30s chenjiel/iq-imatrix-calibration
chenjiel/iq-imatrix-calibration
33m 30s
Upload lm_eval results for qwen 3 8B (mmlu + wikitext) Unit tests #13413: Pull request #2763 opened by danielkorzekwa
18m 30s dkorzekws/hf_qad_nel_results
dkorzekws/hf_qad_nel_results
18m 30s
Support a new model (NemotronH_Omni_Reasoning_V3) in Megatron-Bridge … Unit tests #13412: Commit 1e56773 pushed by yueshen2016
15m 40s main
main
15m 40s