Skip to content

GitLab

Switch branch/tag
  1. 02 Feb, 2025 1 commit
  3. 01 Feb, 2025 4 commits
    • Michael Goin's avatar
      Apply torch.compile to fused_moe/grouped_topk (#12637) · 3194039c
      Michael Goin authored
      3194039c
    • Lucas Wilkinson's avatar
      [Attention] Deepseek v3 MLA support with FP8 compute (#12601) · baeded25
      Lucas Wilkinson authored 
      

This PR implements the Deepseek V3 support by performing matrix absorption the fp8 weights 

---------
Signed-off-by: default avatarLucas Wilkinson <lwilkinson@neuralmagic.com>
Co-authored-by: default avatarWoosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: default avatarsimon-mo <simon.mo@hey.com>
Co-authored-by: default avatarMichael Goin <mgoin64@gmail.com>
Co-authored-by: default avatarZhuohan Li <zhuohan123@gmail.com>
Co-authored-by: default avatarTyler Michael Smith <tysmith@redhat.com>
Co-authored-by: default avatarAlexander Matveev <59768536+alexm-neuralmagic@users.noreply.github.com>
      baeded25
    • Rahul Tuli's avatar
      Fix: Respect `sparsity_config.ignore` in Cutlass Integration (#12517) · 3e1c76cf
      Rahul Tuli authored 
      This PR addresses a bug in the Cutlass integration where the
`sparsity_config.ignore` list was not being respected. When only a
subset of modules were configured as Sparse24, the system incorrectly
selected Cutlass for non-sparse modules as well. This update ensures the
correct scheme is selected for non-sparse modules, fixing this behavior.

---

### Changes

- Updated logic to correctly respect `sparsity_config.ignore`.
- Ensured non-sparse modules use the appropriate scheme instead of
defaulting to Cutlass.

---

<details>
<summary>Testing Setup</summary>

The fix has been tested on top of [this
diff](https://github.com/vllm-project/vllm/pull/12097).

#### Steps to Test:
```bash
git checkout -b my-test-branch origin/rahul-bitmask-additions # compressed Cutlass support
git revert --no-edit aa2cd2c4 # revert Tyler's commit to turn off Cutlass for W16A16
git cherry-pick ca624cddb # this branch
```

#### Additional Patch Required:
```diff
diff --git a/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py b/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py
index a54177c1c..f916dd0c9 100644
--- a/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py
+++ b/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py
@@ -9,7 +9,7 @@ from compressed_tensors.quantization import (QuantizationArgs,
                                              QuantizationStrategy,
                                              QuantizationType)
 from pydantic import BaseModel
-
+from vllm.logger import init_logger
 from vllm.model_executor.layers.fused_moe import FusedMoE
 from vllm.model_executor.layers.linear import (LinearBase, LinearMethodBase,
                                                UnquantizedLinearMethod)
@@ -27,7 +27,7 @@ from vllm.model_executor.layers.quantization.compressed_tensors.utils import (
     should_ignore_layer)
 from vllm.model_executor.layers.quantization.kv_cache import BaseKVCacheMethod
 from vllm.platforms import current_platform
-
+logger = init_logger(__name__)
 __all__ = ["CompressedTensorsLinearMethod"]
 
 SPARSITY_CONFIG_NAME: Literal["sparsity_config"] = "sparsity_config"
```

Apply using:
```bash
git apply logging-patch.patch
```

</details>

---

<details>
<summary>Models Tested</summary>

- `nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-partial-24` 
- `nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-full-sparse24`
-
`nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-partial-24-entire-fp8-compressed`
-
`nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-partial-24-remaining-fp8-compressed`

</details>

---


<details>
<summary>Example Output</summary>

#### Layers 0-5 (Sparse24)
```
Using scheme: CompressedTensors24 for model.layers.0.self_attn.qkv_proj
Using scheme: CompressedTensors24 for model.layers.0.self_attn.o_proj
Using scheme: CompressedTensors24 for model.layers.0.mlp.gate_up_proj
Using scheme: CompressedTensors24 for model.layers.0.mlp.down_proj
...
```

#### Layers 6+ (Non-Sparse, FP8)
```
Using scheme: CompressedTensorsW8A8Fp8 for model.layers.6.self_attn.qkv_proj
Using scheme: CompressedTensorsW8A8Fp8 for model.layers.6.self_attn.o_proj
Using scheme: CompressedTensorsW8A8Fp8 for model.layers.6.mlp.gate_up_proj
Using scheme: CompressedTensorsW8A8Fp8 for model.layers.6.mlp.down_proj
...
```

</details>

**Note:** Assumed all modules in fused layers such as `QKV_proj` and
`Gate_up_proj` follow the same quantization/pruning scheme.

---

For related tasks using the Asana app for GitHub, refer to [[this
link](https://app.asana.com/0/0/1209227810815160)](https://app.asana.com/0/0/1209227810815160

).
Signed-off-by: default avatarRahul Tuli <rahul@neuralmagic.com>
      3e1c76cf
    • Eldar Kurtic's avatar
      Fix target matching for fused layers with compressed-tensors (#12617) · 1867c258
      Eldar Kurtic authored 
      Without this PR
---------------
Quantizing models with llm-compressor and a recipe that explicitly lists
names of layers produces a model that is not loadable by vLLM (i.e.
`vllm serve <model>` fails with `raise ValueError(f"Unable to find
matching target for {module} in the ...`).

Example recipe:
```
recipe = """
quantization_stage:
  run_type: oneshot
  quantization_modifiers:
    GPTQModifier:
      ignore: ["lm_head"]
      config_groups:
        group_0:
          weights:
            num_bits: 4
            type: "int"
            symmetric: true
            strategy: "group"
            group_size: 128
          targets: [
            "model.layers.0.mlp.down_proj",
            "model.layers.2.mlp.down_proj",
            "model.layers.3.mlp.down_proj",
            "model.layers.4.mlp.down_proj",
            "model.layers.5.mlp.down_proj",
            "model.layers.6.mlp.down_proj",
            "model.layers.7.mlp.down_proj",
            "model.layers.8.mlp.down_proj",
            "model.layers.9.mlp.down_proj",
            "model.layers.10.mlp.down_proj",
            "model.layers.11.mlp.down_proj",
            "model.layers.12.mlp.down_proj",
            "model.layers.13.mlp.down_proj",
            "model.layers.14.mlp.down_proj",
            "model.layers.15.mlp.down_proj",
            "model.layers.16.mlp.down_proj",
            "model.layers.17.mlp.down_proj",
            "model.layers.19.mlp.down_proj",
            "model.layers.21.mlp.down_proj",
            "model.layers.22.mlp.down_proj",
            .
            .
            .
          ]
"""
```

To reproduce the vLLM error: 
```bash
vllm serve nm-testing/eldar-test
```

With this PR
------------
Models are loaded correctly without any errors.
      1867c258
  5. 31 Jan, 2025 6 commits
  7. 30 Jan, 2025 1 commit
  9. 29 Jan, 2025 5 commits
  11. 28 Jan, 2025 2 commits
  13. 27 Jan, 2025 3 commits
  15. 26 Jan, 2025 1 commit
  17. 25 Jan, 2025 2 commits
  19. 24 Jan, 2025 1 commit
  21. 23 Jan, 2025 4 commits
  23. 22 Jan, 2025 7 commits
  25. 21 Jan, 2025 3 commits