diff --git a/.github/workflows/zig-build.yml b/.github/workflows/zig-build.yml
new file mode 100644
index 000000000..68a698ab9
--- /dev/null
+++ b/.github/workflows/zig-build.yml
@@ -0,0 +1,25 @@
+name: Zig CI
+
+on:
+  pull_request:
+  push:
+    branches:
+      - master
+
+jobs:
+  build:
+    strategy:
+      fail-fast: false
+      matrix:
+        runs-on: [ubuntu-latest, macos-latest, windows-latest]
+    runs-on: ${{ matrix.runs-on }}
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          submodules: recursive
+          fetch-depth: 0
+      - uses: goto-bus-stop/setup-zig@v2
+        with:
+          version: 0.11.0
+      - name: Build Summary
+        run: zig build --summary all -freference-trace
diff --git a/examples/server/api_like_OAI.py b/examples/server/api_like_OAI.py
index ed19237b0..14d2dcf65 100755
--- a/examples/server/api_like_OAI.py
+++ b/examples/server/api_like_OAI.py
@@ -27,10 +27,10 @@ def is_present(json, key):
         buf = json[key]
     except KeyError:
         return False
+    if json[key] == None:
+        return False
     return True
 
-
-
 #convert chat to prompt
 def convert_chat(messages):
     prompt = "" + args.chat_prompt.replace("\\n", "\n")
diff --git a/ggml-metal.m b/ggml-metal.m
index 8549fa9ac..7ff131cc4 100644
--- a/ggml-metal.m
+++ b/ggml-metal.m
@@ -301,12 +301,11 @@ struct ggml_metal_context * ggml_metal_init(int n_cb) {
 #if TARGET_OS_OSX
     // print MTL GPU family:
     GGML_METAL_LOG_INFO("%s: GPU name:   %s\n", __func__, [[ctx->device name] UTF8String]);
-    GGML_METAL_LOG_INFO("%s: GPU arch:   %s\n", __func__, [[ctx->device architecture].name UTF8String]);
 
     // determine max supported GPU family
     // https://developer.apple.com/metal/Metal-Shading-Language-Specification.pdf
     // https://developer.apple.com/metal/Metal-Feature-Set-Tables.pdf
-    for (int i = MTLGPUFamilyApple9 + 10; i >= MTLGPUFamilyApple1; --i) {
+    for (int i = MTLGPUFamilyApple1 + 20; i >= MTLGPUFamilyApple1; --i) {
         if ([ctx->device supportsFamily:i]) {
             GGML_METAL_LOG_INFO("%s: GPU family: MTLGPUFamilyApple%d (%d)\n", __func__, i - MTLGPUFamilyApple1 + 1, i);
             break;
diff --git a/k_quants.h b/k_quants.h
index adc6a3913..9de089e7a 100644
--- a/k_quants.h
+++ b/k_quants.h
@@ -29,7 +29,7 @@
 
 // 2-bit quantization
 // weight is represented as x = a * q + b
-// 16 blocks of 16 elemenets each
+// 16 blocks of 16 elements each
 // Effectively 2.5625 bits per weight
 typedef struct {
     uint8_t scales[QK_K/16]; // scales and mins, quantized with 4 bits
@@ -41,7 +41,7 @@ static_assert(sizeof(block_q2_K) == 2*sizeof(ggml_fp16_t) + QK_K/16 + QK_K/4, "w
 
 // 3-bit quantization
 // weight is represented as x = a * q
-// 16 blocks of 16 elemenets each
+// 16 blocks of 16 elements each
 // Effectively 3.4375 bits per weight
 #ifdef GGML_QKK_64
 typedef struct {
@@ -62,7 +62,7 @@ static_assert(sizeof(block_q3_K) == sizeof(ggml_fp16_t) + QK_K / 4 + QK_K / 8 +
 #endif
 
 // 4-bit quantization
-// 16 blocks of 32 elements each
+// 8 blocks of 32 elements each
 // weight is represented as x = a * q + b
 // Effectively 4.5 bits per weight
 #ifdef GGML_QKK_64
@@ -83,7 +83,7 @@ static_assert(sizeof(block_q4_K) == 2*sizeof(ggml_fp16_t) + K_SCALE_SIZE + QK_K/
 #endif
 
 // 5-bit quantization
-// 16 blocks of 32 elements each
+// 8 blocks of 32 elements each
 // weight is represented as x = a * q + b
 // Effectively 5.5 bits per weight
 #ifdef GGML_QKK_64
@@ -107,7 +107,7 @@ static_assert(sizeof(block_q5_K) == 2*sizeof(ggml_fp16_t) + K_SCALE_SIZE + QK_K/
 
 // 6-bit quantization
 // weight is represented as x = a * q
-// 16 blocks of 16 elemenets each
+// 16 blocks of 16 elements each
 // Effectively 6.5625 bits per weight
 typedef struct {
     uint8_t ql[QK_K/2];      // quants, lower 4 bits