Merge branch 'main' into claude/nervous-meitner

[CI] Enable Eagle2.5 in basic model test CI
Remove is_available_online=False for Eagle2_5_VLForConditionalGeneration as nvidia/Eagle2.5-8B is publicly available on HuggingFace. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: mgoin <mgoin64@gmail.com>
2026-03-17 17:39:29 +01:00 · 2026-03-17 12:07:19 -04:00 · 2026-03-17 11:58:52 -04:00
1 changed files with 5 additions and 7 deletions
@@ -286,7 +286,7 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
        min_transformers_version="5.0.0",
    ),
    "GlmMoeDsaForCausalLM": _HfExamplesInfo(
-        "zai-org/GLM-5", min_transformers_version="5.0.1", is_available_online=False
+        "zai-org/GLM-5", min_transformers_version="5.0.1"
    ),
    "GPT2LMHeadModel": _HfExamplesInfo("openai-community/gpt2", {"alias": "gpt2"}),
    "GPTBigCodeForCausalLM": _HfExamplesInfo(
@@ -774,7 +774,7 @@ _MULTIMODAL_EXAMPLE_MODELS = {
        "rednote-hilab/dots.ocr", trust_remote_code=True
    ),
    "Eagle2_5_VLForConditionalGeneration": _HfExamplesInfo(
-        "nvidia/Eagle2.5-8B", trust_remote_code=True, is_available_online=False
+        "nvidia/Eagle2.5-8B", trust_remote_code=True
    ),
    "Emu3ForConditionalGeneration": _HfExamplesInfo("BAAI/Emu3-Chat-hf"),
    "Ernie4_5_VLMoeForConditionalGeneration": _HfExamplesInfo(
@@ -971,7 +971,7 @@ _MULTIMODAL_EXAMPLE_MODELS = {
        trust_remote_code=True,
    ),
    "NemotronH_Nano_VL_V2": _HfExamplesInfo(
-        "nano_vl_dummy", is_available_online=False, trust_remote_code=True
+        "nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16", trust_remote_code=True
    ),
    "OpenCUAForConditionalGeneration": _HfExamplesInfo(
        "xlangai/OpenCUA-7B", trust_remote_code=True
@@ -1252,9 +1252,8 @@ _SPECULATIVE_DECODING_EXAMPLE_MODELS = {
        speculative_model="XiaomiMiMo/MiMo-7B-RL",
    ),
    "NemotronHMTPModel": _HfExamplesInfo(
-        "nvidia/Nemotron-Super-Placeholder",
-        speculative_model="nvidia/Nemotron-Super-Placeholder",
-        is_available_online=False,
+        "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
+        speculative_model="nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16",
    ),
    "OpenPanguMTPModel": _HfExamplesInfo(
        "FreedomIntelligence/openPangu-Ultra-MoE-718B-V1.1",
@@ -1278,7 +1277,6 @@ _SPECULATIVE_DECODING_EXAMPLE_MODELS = {
        use_original_num_layers=True,
        # Initialize at least one MoE layer
        hf_overrides={"num_hidden_layers": 4},
-        is_available_online=False,
    ),
 }