From a46abb7ae68acc13a4fc5870db98619b3f97c6e0 Mon Sep 17 00:00:00 2001 From: Ting SUN Date: Wed, 17 Jun 2026 13:08:41 +0800 Subject: [PATCH] [Bugfix][Quantization] Reject unsupported compressed tensors KV cache schemes (#45312) Signed-off-by: Ting Sun Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com> --- .../quantization/compressed_tensors/compressed_tensors.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) 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 71fc809f4d..2231b2ca9a 100644 --- a/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py +++ b/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py @@ -983,7 +983,7 @@ class CompressedTensorsKVCacheMethod(BaseKVCacheMethod): type_ = kv_cache_scheme.get("type") num_bits = kv_cache_scheme.get("num_bits") - if type_ != "float" and num_bits != 8: + if type_ != "float" or num_bits != 8: raise NotImplementedError( "Currently supported kv cache quantization is " "num_bits=8, type=float, however "