From 2dee7cf96479ea7cfc50dc5bb0520cba6a51aaca Mon Sep 17 00:00:00 2001
From: Younes B <49240599+younesbelkada@users.noreply.github.com>
Date: Tue, 8 Jul 2025 10:43:50 +0400
Subject: [PATCH 1/2] Apply suggestions from code review

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
---
 gguf-py/gguf/constants.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py
index 84a91e82ca..5ae4eee802 100644
--- a/gguf-py/gguf/constants.py
+++ b/gguf-py/gguf/constants.py
@@ -289,7 +289,7 @@ class MODEL_ARCH(IntEnum):
     LLAMA4           = auto()
     DECI             = auto()
     FALCON           = auto()
-    FALCON_H1           = auto()
+    FALCON_H1        = auto()
     BAICHUAN         = auto()
     GROK             = auto()
     GPT2             = auto()
@@ -662,7 +662,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
     MODEL_ARCH.DOTS1:            "dots1",
     MODEL_ARCH.ARCEE:            "arcee",
     MODEL_ARCH.ERNIE4_5:         "ernie4_5",
-    MODEL_ARCH.FALCON_H1:         "falcon_h1",
+    MODEL_ARCH.FALCON_H1:        "falcon_h1",
 }
 
 VISION_PROJECTOR_TYPE_NAMES: dict[VISION_PROJECTOR_TYPE, str] = {

From a846d02327b08e58528d5967ad5cfe3f30f2087a Mon Sep 17 00:00:00 2001
From: younesbelkada <younes.belkada@tii.ae>
Date: Tue, 8 Jul 2025 10:44:59 +0400
Subject: [PATCH 2/2] remove todo

---
 gguf-py/gguf/tensor_mapping.py | 5 +----
 src/llama-arch.cpp             | 3 +--
 2 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/gguf-py/gguf/tensor_mapping.py b/gguf-py/gguf/tensor_mapping.py
index ff3f273bd5..b8b081a90b 100644
--- a/gguf-py/gguf/tensor_mapping.py
+++ b/gguf-py/gguf/tensor_mapping.py
@@ -583,6 +583,7 @@ class TensorNameMap:
         ),
 
         MODEL_TENSOR.SSM_NORM: (
+            "model.layers.{bid}.mamba.norm", # falcon-h1
             "backbone.layers.{bid}.mixer.norm",  # mamba2
         ),
 
@@ -1177,10 +1178,6 @@ class TensorNameMap:
             "resampler.attn.out_proj",
         ),
         
-        MODEL_TENSOR.SSM_NORM: (
-            "model.layers.{bid}.mamba.norm",
-        ),
-
         MODEL_TENSOR.V_RESMPL_KV: (
             "resampler.kv_proj",
         ),
diff --git a/src/llama-arch.cpp b/src/llama-arch.cpp
index b6bd815090..f6c092cf20 100644
--- a/src/llama-arch.cpp
+++ b/src/llama-arch.cpp
@@ -1955,8 +1955,7 @@ bool llm_arch_is_recurrent(const llm_arch & arch) {
 }
 
 bool llm_arch_is_hybrid(const llm_arch & arch) {
-    // TODO: There are currently no hybrid models! Once there are, this will be
-    //  the place to identify them
+    // List all mamba-attention hybrid models here
     switch (arch) {
         case LLM_ARCH_FALCON_H1:
             return true;