Add IQ2_KL

README.md

@@ -298,10 +298,8 @@ custom=$(
 </details>
 
 
-## `IQ2_KL` 10.913 GiB (3.070 BPW)
-Final estimate: PPL = 7.7567 +/- 0.05475
-
-*NOTE* this test recipe may change before uploading!
+## `IQ2_KL` 11.516 GiB (3.240 BPW)
+Final estimate: PPL = 7.7121 +/- 0.05402
 
 <details>
 
@@ -314,14 +312,20 @@ custom="
 # 48 Repeating Layers [0-47]
 
 # Attention
+blk\.(0)\.attn_q.*=q8_0
+blk\.(0)\.attn_k.*=q8_0
+blk\.(0)\.attn_v.*=q8_0
+blk\.(0)\.attn_output.*=q8_0
+
 blk\..*\.attn_q.*=iq5_k
 blk\..*\.attn_k.*=iq6_k
 blk\..*\.attn_v.*=iq6_k
 blk\..*\.attn_output.*=iq5_k
 
 # Routed Experts
-blk\.(0|1|2|47)\.ffn_down_exps\.weight=iq4_ks
-blk\.(0|1|2|47)\.ffn_(gate|up)_exps\.weight=iq3_ks
+blk\.(0|47)\.ffn_down_exps\.weight=q8_0
+blk\.(0|47)\.ffn_(gate|up)_exps\.weight=q8_0
+
 blk\..*\.ffn_down_exps\.weight=iq3_ks
 blk\..*\.ffn_(gate|up)_exps\.weight=iq2_kl
 
@@ -335,10 +339,9 @@ custom=$(
   sed -Ez 's:\n+:,:g;s:,$::;s:^,::'
 )
 
-    #--imatrix /mnt/raid/models/ubergarm/Qwen3-30B-A3B-Instruct-2507-GGUF/imatrix-Qwen3-30B-A3B-Instruct-2507-BF16.dat \
 ./build/bin/llama-quantize \
     --custom-q "$custom" \
-    --imatrix /mnt/raid/models/ubergarm/Qwen3-30B-A3B-Instruct-2507-GGUF/imatrix-Qwen3-30B-A3B-Instruct-2507-BF16.dat \
+    --imatrix /mnt/raid/models/ubergarm/Qwen3-30B-A3B-Instruct-2507-GGUF/imatrix-eaddario-combined-all-medium-Qwen3-30B-A3B-Instruct-2507-BF16.dat \
     /mnt/raid/models/ubergarm/Qwen3-30B-A3B-Instruct-2507-GGUF/Qwen3-30B-A3B-Instruct-2507-BF16-00001-of-00002.gguf \
     /mnt/raid/models/ubergarm/Qwen3-30B-A3B-Instruct-2507-GGUF/Qwen3-30B-A3B-Instruct-2507-IQ2_KL.gguf \
     IQ2_KL \