Update matmul op args & Update test_mla_decoding_split_kv parameter & other updates

modeling/transformers/pyproject.toml

@@ -60,6 +60,7 @@ tilegym_hf_bench = ["kernel_filters/*.yaml"]
 managed = true
 override-dependencies = [
   "transformers==5.10.2",
+  "triton==3.7.1",
   "cuda-toolkit[tileiras,nvcc,nvvm]==13.3.0",
   "nvidia-cuda-runtime>=13.3,<14",
   "nvidia-nvjitlink>=13.3,<14",

skills/tilegym-monkey-patch-kernels-to-transformers/BENCHMARK.md

@@ -7,7 +7,7 @@ This benchmark summarizes 3-Tier Evaluation from NVSkills-Eval results for the s
 ## Evaluation Summary
 
 - Skill: `tilegym-monkey-patch-kernels-to-transformers`
-- Evaluation date: 2026-06-16
+- Evaluation date: 2026-07-10
 - NVSkills-Eval profile: `external`
 - Environment: `astra-sandbox`
 - Dataset: 5 evaluation tasks
@@ -55,10 +55,10 @@ Task composition is derived from the evaluation dataset when possible. Entries w
 | Dimension | Num | `claude-code` | `codex` |
 |---|---:|---:|---:|
 | Security | 5 | 100% (+0%) | 100% (+0%) |
-| Correctness | 5 | 100% (+12%) | 99% (+12%) |
-| Discoverability | 5 | 100% (+11%) | 94% (+2%) |
-| Effectiveness | 5 | 98% (+18%) | 100% (+19%) |
-| Efficiency | 5 | 96% (+13%) | 90% (+1%) |
+| Correctness | 5 | 100% (+15%) | 99% (+14%) |
+| Discoverability | 5 | 100% (+13%) | 99% (+9%) |
+| Effectiveness | 5 | 98% (+17%) | 100% (+19%) |
+| Efficiency | 5 | 96% (+13%) | 96% (+7%) |
 
 Score values show skill-assisted performance. Values in parentheses show uplift versus the no-skill baseline when baseline data is available.
 

skills/tilegym-monkey-patch-kernels-to-transformers/references/kernel-inventory-schema.md

@@ -52,10 +52,10 @@ Use FlashInfer Solution metadata with source paths:
 - `name`, `definition`, `author`, `spec`, and `sources` are required.
 - `spec.language` is `cuda-tile` for cuTile kernels.
 - `spec.entry_point` uses `{file_path}::{function_name}` and points at `kernels/<kernel_name>.py`.
-- `spec.target_hardware` lists supported GPUs, for example `NVIDIA_B200`.
+- `spec.target_hardware` lists supported CUDA compute capabilities, for example `SM100`.
 - `sources.path` references one or more repo-relative files containing the implementation.
 
-For Ocean in-repo inventory, `sources.content` is not required. If an external FlashInfer-Bench submission needs embedded file content, materialize it from `sources.path` at export time.
+For TileGym in-repo inventory, `sources.content` is not required. If an external FlashInfer-Bench submission needs embedded file content, materialize it from `sources.path` at export time.
 
 ## Agent workflow rules
 

skills/tilegym-monkey-patch-kernels-to-transformers/skill-card.md

@@ -9,11 +9,17 @@ NVIDIA <br>
 ### License/Terms of Use: <br>
 CC-BY-4.0 AND Apache-2.0 <br>
 ## Use Case: <br>
-Developers and engineers integrating TileGym GPU kernels into Hugging Face transformers models for LLM training and inference performance improvements. <br>
+Developers and engineers integrating TileGym high-performance CUDA Tile kernels into Hugging Face transformers models to validate end-to-end functional correctness and improve inference and training throughput. <br>
 
 ### Deployment Geography for Use: <br>
 Global <br>
 
+## Requirements / Dependencies: <br>
+**Requires API Key or External Credential:** [Not Specified] <br>
+**Credential Type(s):** [None identified] <br>
+
+Do not include secrets in prompts/logs/output; use least-privilege credentials; rotate keys as appropriate. <br>
+
 ## Known Risks and Mitigations: <br>
 Risk: Review before execution as proposals could introduce incorrect or misleading guidance into skills. <br>
 Mitigation: Review and scan skill before deployment. <br>
@@ -23,11 +29,14 @@ Mitigation: Review and scan skill before deployment. <br>
 - [Kernel Integration](references/kernel-integration.md) <br>
 - [Auto Kernelize](references/auto-kernelize.md) <br>
 - [Kernel Inventory Schema](references/kernel-inventory-schema.md) <br>
-- [NVIDIA CUDA Tile IR Documentation](https://docs.nvidia.com/cuda/tile-ir/latest/) <br>
+- [Workflow Diagram](references/workflow-diagram.png) <br>
+- [CUDA Tile IR Documentation](https://docs.nvidia.com/cuda/tile-ir/latest/) <br>
+- [FlashInfer-Bench Definition Schema](https://github.com/flashinfer-ai/flashinfer-bench/blob/main/docs/flashinfer-trace/definition.mdx) <br>
+- [FlashInfer-Bench Solution Schema](https://github.com/flashinfer-ai/flashinfer-bench/blob/main/docs/flashinfer-trace/solution.mdx) <br>
 
 
 ## Skill Output: <br>
-**Output Type(s):** [Code, Shell commands, Configuration instructions] <br>
+**Output Type(s):** [Code, Shell commands, Analysis] <br>
 **Output Format:** [Markdown with inline code blocks] <br>
 **Output Parameters:** [1D] <br>
 **Other Properties Related to Output:** [None] <br>
@@ -39,7 +48,7 @@ Mitigation: Review and scan skill before deployment. <br>
 
 
 ## Evaluation Tasks: <br>
-Evaluated against 5 tasks (1 positive skill-activation, 4 negative) in the NVSkills-Eval `external` profile. <br>
+Evaluated against 5 tasks (1 positive skill-activation, 4 negative) via NVSkills-Eval external profile in astra-sandbox environment. <br>
 
 ## Evaluation Metrics Used: <br>
 Reported benchmark dimensions: <br>
@@ -64,10 +73,10 @@ Underlying evaluation signals used in this run: <br>
 | Dimension | Num | `claude-code` | `codex` |
 |---|---:|---:|---:|
 | Security | 5 | 100% (+0%) | 100% (+0%) |
-| Correctness | 5 | 100% (+12%) | 99% (+12%) |
-| Discoverability | 5 | 100% (+11%) | 94% (+2%) |
-| Effectiveness | 5 | 98% (+18%) | 100% (+19%) |
-| Efficiency | 5 | 96% (+13%) | 90% (+1%) |
+| Correctness | 5 | 100% (+15%) | 99% (+14%) |
+| Discoverability | 5 | 100% (+13%) | 99% (+9%) |
+| Effectiveness | 5 | 98% (+17%) | 100% (+19%) |
+| Efficiency | 5 | 96% (+13%) | 96% (+7%) |
 
 ## Skill Version(s): <br>
 2026.06.03 (source: frontmatter) <br>

skills/tilegym-monkey-patch-kernels-to-transformers/skill.oms.sig

@@ -1 +1 @@
-{"mediaType":"application/vnd.dev.sigstore.bundle.v0.3+json","verificationMaterial":{"x509CertificateChain":{"certificates":[{"rawBytes":"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"},{"rawBytes":"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"},{"rawBytes":"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"}]},"tlogEntries":[]},"dsseEnvelope":{"payload":"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","payloadType":"application/vnd.in-toto+json","signatures":[{"sig":"MGUCMQDbfR92FeMjEcZQzIbdqnlmgaxCNd7nGVXKp6/pladsoMnT3ckTYOfsHF6MxAi9+5ECME668JO/ltcRgtzmnMcr2y+8p56SBhMcuQt7UYx6w8Z+9ShLyaZHF6+3gK7hFd6LFw==","keyid":""}]}}
+{"mediaType":"application/vnd.dev.sigstore.bundle.v0.3+json","verificationMaterial":{"x509CertificateChain":{"certificates":[{"rawBytes":"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"},{"rawBytes":"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"},{"rawBytes":"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"}]},"tlogEntries":[]},"dsseEnvelope":{"payload":"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","payloadType":"application/vnd.in-toto+json","signatures":[{"sig":"MGQCMDWygNgixRS5Om5qKqmF0KeGegPpXb3Hqk6XlkrTnKM7l+9zOLmhYcyML1ncceiL9AIwVNBfYl9mWsMkkd1fNSALxLytTGTtohgh1wSZwPhZCgKwMRWJL7aX5dwl7F8E7n9R","keyid":""}]}}

src/tilegym/__init__.py

@@ -17,19 +17,6 @@ def _check_torch_dependencies():
         ) from None
 
 
-def _check_ct_tune_dependency():
-    """Verify that cuda-tile with tune support is installed with helpful error message."""
-    try:
-        import cuda.tile.tune  # noqa: F401
-    except (ImportError, ModuleNotFoundError):
-        raise ImportError(
-            "\n\n[TileGym] cuda.tile.tune is required but not available.\n"
-            "Please install or upgrade cuda-tile:\n\n"
-            "  pip install cuda-tile\n\n"
-            "See: https://github.com/NVIDIA/cutile-python"
-        ) from None
-
-
 # Check dependencies before any imports
 _check_torch_dependencies()
 
@@ -54,7 +41,6 @@ def _check_ct_tune_dependency():
 
 # Setup cutile integration
 if is_backend_available("cutile"):
-    _check_ct_tune_dependency()
     # Apply experimental kernel tracking patch
     from .experimental import _apply_patch as _apply_experimental_patch
 

src/tilegym/backend/selector.py

@@ -29,17 +29,25 @@ def is_nvt_available():
 
 try:
     import cuda.tile as ct
+    import cuda.tile.tune  # noqa: F401  # required by every op under ops/cutile/
 
     CUTILE_AVAILABLE = True
 
 except ImportError:
     import warnings
 
-    warnings.warn("Failed to import cuda_tile_compiler, CUDA Tile backend is not available")
+    warnings.warn(
+        "Failed to import cuda.tile / cuda.tile.tune, CUDA Tile backend is not available. "
+        "To enable it: `pip install cuda-tile` "
+        "(see https://github.com/NVIDIA/cutile-python)"
+    )
     CUTILE_AVAILABLE = False
 
 
 def is_cutile_available():
+    if os.environ.get("TILEGYM_DISABLE_CUTILE") == "1":
+        print("[TileGym] TILEGYM_DISABLE_CUTILE=1; CUDA Tile backend force-disabled")
+        return False
     return CUTILE_AVAILABLE