diff --git a/docs/articles_en/about-openvino/performance-benchmarks.rst b/docs/articles_en/about-openvino/performance-benchmarks.rst
index 233067735eae43..5d414ab4909ecb 100644
--- a/docs/articles_en/about-openvino/performance-benchmarks.rst
+++ b/docs/articles_en/about-openvino/performance-benchmarks.rst
@@ -160,7 +160,7 @@ For a listing of all platforms and configurations used for testing, refer to the
**Disclaimers**
* Intel® Distribution of OpenVINO™ toolkit performance results are based on release
- 2025.0, as of February 05, 2025.
+ 2025.0 as of January 28, 2025.
* OpenVINO Model Server performance results are based on release
2024.5, as of November 20, 2024.
diff --git a/docs/articles_en/about-openvino/performance-benchmarks/model-accuracy-int8-fp32.rst b/docs/articles_en/about-openvino/performance-benchmarks/model-accuracy-int8-fp32.rst
index cc702d5eddec73..1163fced6220a5 100644
--- a/docs/articles_en/about-openvino/performance-benchmarks/model-accuracy-int8-fp32.rst
+++ b/docs/articles_en/about-openvino/performance-benchmarks/model-accuracy-int8-fp32.rst
@@ -66,6 +66,7 @@ the table for more information.
- -0.09%
- -0.02%
- -0.04%
+
.. list-table:: Model Accuracy for BF16, FP32 and FP16 (FP16: Arc only. BF16: Xeon® 6972P only)
:header-rows: 1
@@ -125,7 +126,8 @@ the table for more information.
- 0.01%
-
- -0.03%
-.. list-table:: Model Accuracy for AMX-FP16, AMX-INT4, Arc-FP16 and Arc-INT4 (Arc™ A-series)
+
+.. list-table:: Model Accuracy for AMX-FP16, AMX-INT4, Arc-FP16 and Arc-INT4 (Arc™ B-series)
:header-rows: 1
* - OpenVINO™ Model name
@@ -135,6 +137,13 @@ the table for more information.
- B, AMX-INT4
- C, Arc-FP16
- D, Arc-INT4
+ * - GLM4-9B-Chat
+ - Data Default WWB
+ - Similarity
+ - 6.9%
+ - 3.8%
+ - 6.3%
+ - 15.1%
* - Qwen-2.5-7B-instruct
- Data Default WWB
- Similarity
@@ -209,6 +218,7 @@ the table for more information.
Notes: For all accuracy metrics a "-", (minus sign), indicates an accuracy drop.
The Similarity metric is the distance from "perfect" and as such always positive.
Similarity is cosine similarity - the dot product of two vectors divided by the product of their lengths.
+
.. raw:: html
diff --git a/docs/articles_en/about-openvino/performance-benchmarks/performance-benchmarks-faq.rst b/docs/articles_en/about-openvino/performance-benchmarks/performance-benchmarks-faq.rst
index 9d5f92dbadbad2..00815e9f1ffecd 100644
--- a/docs/articles_en/about-openvino/performance-benchmarks/performance-benchmarks-faq.rst
+++ b/docs/articles_en/about-openvino/performance-benchmarks/performance-benchmarks-faq.rst
@@ -63,15 +63,15 @@ Performance Information F.A.Q.
- Meta AI
- Auto regressive language
- 8K
- * - `Llama-3.2-3B `__
+ * - `Llama-3.2-3B-Instruct `__
- Meta AI
- Auto regressive language
- 128K
- * - `Mistral-7b-V0.1 `__
+ * - `Mistral-7b-Instruct-V0.2 `__
- Mistral AI
- Auto regressive language
- - 4096
- * - `Phi3-4k-mini `__
+ - 32K
+ * - `Phi3-4k-mini-Instruct `__
- Huggingface
- Auto regressive language
- 4096
@@ -79,20 +79,43 @@ Performance Information F.A.Q.
- Huggingface
- Auto regressive language
- 128K
+ * - `Qwen-2.5-7B-Instruct `__
+ - Huggingface
+ - Auto regressive language
+ - 128K
* - `Stable-Diffusion-V1-5 `__
- Hugginface
- Latent Diffusion Model
- 77
+ * - `FLUX.1-schnell `__
+ - Hugginface
+ - Latent Adversarial Diffusion Distillation Model
+ - 256
* - `bert-base-cased `__
- BERT
- question / answer
- 128
+ * - `mask_rcnn_resnet50_atrous_coco `__
+ - Mask R-CNN ResNet 50 Atrous
+ - object instance segmentation
+ - 800x1365
+ * - `mobilenet-v2 `__
+ - Mobilenet V2 PyTorch
+ - classification
+ - 224x224
+ * - `resnet-50 `__
+ - ResNet-50_v1_ILSVRC-2012
+ - classification
+ - 224x224
+ * - `ssd-resnet34-1200-onnx `__
+ - ssd-resnet34 onnx model
+ - object detection
+ - 1200x1200
* - `yolov8n `__
- Yolov8nano
- object detection
- 608x608
-
.. dropdown:: Where can I purchase the specific hardware used in the benchmarking?
Intel partners with vendors all over the world. For a list of Hardware Manufacturers, see the
diff --git a/docs/sphinx_setup/_static/benchmarks_files/data/graph-data-ov.json b/docs/sphinx_setup/_static/benchmarks_files/data/graph-data-ov.json
index 9b765d15b1a8b3..55919d39e6cacb 100644
--- a/docs/sphinx_setup/_static/benchmarks_files/data/graph-data-ov.json
+++ b/docs/sphinx_setup/_static/benchmarks_files/data/graph-data-ov.json
@@ -10,7 +10,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.04,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -220,7 +220,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.04,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -233,7 +233,7 @@
"Precisions": [
{
"int4": "",
- "int8": 20456.96,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -501,7 +501,7 @@
{
"int4": "",
"int8": "",
- "fp16": 2.83,
+ "fp16": "",
"fp32": "",
"bf16": ""
}
@@ -799,7 +799,7 @@
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -955,7 +955,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1.78,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -968,13 +968,13 @@
"Precisions": [
{
"int4": "",
- "int8": 561.11,
+ "int8": 16.0,
"fp16": "",
"fp32": "",
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -1429,42 +1429,7 @@
"bf16": ""
}
],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Arc™ A-Series Graphics dGPU",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": true,
- "whats_new_model": false,
- "PlatformType": "Accelerator Platforms",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 2.18,
- "fp16": 2.19,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -1865,7 +1830,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.14,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -1878,7 +1843,7 @@
"Precisions": [
{
"int4": "",
- "int8": 6835.33,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -2075,7 +2040,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.06,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -2088,7 +2053,7 @@
"Precisions": [
{
"int4": "",
- "int8": 16196.18,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -2285,9 +2250,9 @@
"Precisions": [
{
"int4": "",
- "int8": 0.18,
+ "int8": "",
"fp16": "",
- "fp32": 0.35,
+ "fp32": "",
"bf16": ""
}
],
@@ -2530,9 +2495,9 @@
"Precisions": [
{
"int4": "",
- "int8": 0.17,
+ "int8": "",
"fp16": "",
- "fp32": 0.04,
+ "fp32": "",
"bf16": ""
}
],
@@ -2543,7 +2508,7 @@
"Precisions": [
{
"int4": "",
- "int8": 5759.21,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -2952,7 +2917,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.87,
+ "fp32": "",
"bf16": ""
}
],
@@ -3197,7 +3162,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.33,
+ "fp32": "",
"bf16": ""
}
],
@@ -4094,41 +4059,6 @@
}
}
},
- {
- "Platform": "Intel® Core™ Ultra 7 processor 155H iGPU-only",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": true,
- "whats_new_model": false,
- "PlatformType": "Intel® Core™, iGPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 11.12,
- "fp16": 10.73,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
{
"Platform": "Intel® Core™ Ultra 7 processor 155H iGPU-only",
"Model": "stable-diffusion-v1-5",
@@ -4159,7 +4089,7 @@
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -4282,7 +4212,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.54,
+ "fp32": "",
"bf16": ""
}
],
@@ -4527,7 +4457,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.23,
+ "fp32": "",
"bf16": ""
}
],
@@ -5191,7 +5121,7 @@
{
"int4": "",
"int8": "",
- "fp16": 0.75,
+ "fp16": "",
"fp32": "",
"bf16": ""
}
@@ -5459,41 +5389,6 @@
}
}
},
- {
- "Platform": "Intel® Core™ Ultra 9 processor 288V iGPU-only",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": true,
- "whats_new_model": false,
- "PlatformType": "Intel® Core™, iGPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 5.46,
- "fp16": 5.61,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
{
"Platform": "Intel® Core™ Ultra 9 processor 288V iGPU-only",
"Model": "stable-diffusion-v1-5",
@@ -5524,7 +5419,7 @@
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -5612,7 +5507,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.15,
+ "fp32": "",
"bf16": ""
}
],
@@ -5855,7 +5750,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.43,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -5868,7 +5763,7 @@
"Precisions": [
{
"int4": "",
- "int8": 2514.0,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -6277,7 +6172,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.22,
+ "fp32": "",
"bf16": ""
}
],
@@ -6520,7 +6415,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.57,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -6533,7 +6428,7 @@
"Precisions": [
{
"int4": "",
- "int8": 2127.48,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -6940,9 +6835,9 @@
"Precisions": [
{
"int4": "",
- "int8": 1.68,
+ "int8": "",
"fp16": "",
- "fp32": 0.49,
+ "fp32": "",
"bf16": ""
}
],
@@ -6953,7 +6848,7 @@
"Precisions": [
{
"int4": "",
- "int8": 734.76,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -7185,9 +7080,9 @@
"Precisions": [
{
"int4": "",
- "int8": 0.59,
+ "int8": "",
"fp16": "",
- "fp32": 0.39,
+ "fp32": "",
"bf16": ""
}
],
@@ -7430,9 +7325,9 @@
"Precisions": [
{
"int4": "",
- "int8": 0.71,
+ "int8": "",
"fp16": "",
- "fp32": 0.19,
+ "fp32": "",
"bf16": ""
}
],
@@ -7443,7 +7338,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1352.54,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -7852,7 +7747,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.14,
+ "fp32": "",
"bf16": ""
}
],
@@ -8095,9 +7990,9 @@
"Precisions": [
{
"int4": "",
- "int8": 0.5,
+ "int8": "",
"fp16": "",
- "fp32": 0.13,
+ "fp32": "",
"bf16": ""
}
],
@@ -8108,7 +8003,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1834.27,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -8515,7 +8410,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1.14,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -8760,9 +8655,9 @@
"Precisions": [
{
"int4": "",
- "int8": 1.35,
+ "int8": "",
"fp16": "",
- "fp32": 0.35,
+ "fp32": "",
"bf16": ""
}
],
@@ -8773,7 +8668,7 @@
"Precisions": [
{
"int4": "",
- "int8": 895.22,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -9145,9 +9040,9 @@
"Precisions": [
{
"int4": "",
- "int8": 0.54,
+ "int8": "",
"fp16": "",
- "fp32": 0.48,
+ "fp32": "",
"bf16": ""
}
],
@@ -9392,7 +9287,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.18,
+ "fp32": "",
"bf16": ""
}
],
@@ -9810,7 +9705,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1.3,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -9823,13 +9718,13 @@
"Precisions": [
{
"int4": "",
- "int8": 765.77,
+ "int8": 36.8,
"fp16": "",
"fp32": "",
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -10254,41 +10149,6 @@
}
}
},
- {
- "Platform": "Intel® Data Center GPU Flex 170 dGPU",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": false,
- "whats_new_model": false,
- "PlatformType": "Accelerator Platforms",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 1.91,
- "fp16": 1.92,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
{
"Platform": "Intel® Data Center GPU Flex 170 dGPU",
"Model": "stable-diffusion-v1-5",
@@ -10319,7 +10179,7 @@
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -10545,7 +10405,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.16,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -10558,7 +10418,7 @@
"Precisions": [
{
"int4": "",
- "int8": 6247.36,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -10768,7 +10628,7 @@
"Precisions": [
{
"int4": "",
- "int8": 21420.36,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -10920,7 +10780,7 @@
}
},
{
- "Platform": "Intel® Xeon® GNR29 CPU-only",
+ "Platform": "Intel® Xeon® Gold 5218T CPU-only",
"Model": "bert-base-cased",
"featured_SKU": false,
"whats_new_model": false,
@@ -10930,9 +10790,9 @@
"Precisions": [
{
"int4": "",
- "int8": "",
+ "int8": 216.65,
"fp16": "",
- "fp32": "",
+ "fp32": 80.03,
"bf16": ""
}
],
@@ -10943,7 +10803,7 @@
"Precisions": [
{
"int4": "",
- "int8": 4.47,
+ "int8": 14.43,
"fp16": "",
"fp32": "",
"bf16": ""
@@ -10955,8 +10815,8 @@
}
},
{
- "Platform": "Intel® Xeon® GNR29 CPU-only",
- "Model": "mobilenet-v2",
+ "Platform": "Intel® Xeon® Gold 5218T CPU-only",
+ "Model": "mask_rcnn_resnet50_atrous_coco",
"featured_SKU": false,
"whats_new_model": false,
"PlatformType": "Intel® Xeon®, CPU-only",
@@ -10978,7 +10838,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1.93,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -10990,8 +10850,8 @@
}
},
{
- "Platform": "Intel® Xeon® GNR29 CPU-only",
- "Model": "resnet-50",
+ "Platform": "Intel® Xeon® Gold 5218T CPU-only",
+ "Model": "mobilenet-v2",
"featured_SKU": false,
"whats_new_model": false,
"PlatformType": "Intel® Xeon®, CPU-only",
@@ -11000,9 +10860,9 @@
"Precisions": [
{
"int4": "",
- "int8": "",
+ "int8": 5447.03,
"fp16": "",
- "fp32": "",
+ "fp32": 1910.7,
"bf16": ""
}
],
@@ -11013,7 +10873,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1.88,
+ "int8": 1.38,
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11025,8 +10885,8 @@
}
},
{
- "Platform": "Intel® Xeon® GNR29 CPU-only",
- "Model": "ssd-resnet34-1200",
+ "Platform": "Intel® Xeon® Gold 5218T CPU-only",
+ "Model": "resnet-50",
"featured_SKU": false,
"whats_new_model": false,
"PlatformType": "Intel® Xeon®, CPU-only",
@@ -11035,9 +10895,9 @@
"Precisions": [
{
"int4": "",
- "int8": "",
+ "int8": 958.08,
"fp16": "",
- "fp32": "",
+ "fp32": 269.68,
"bf16": ""
}
],
@@ -11048,7 +10908,7 @@
"Precisions": [
{
"int4": "",
- "int8": 9.35,
+ "int8": 3.08,
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11060,8 +10920,8 @@
}
},
{
- "Platform": "Intel® Xeon® GNR29 CPU-only",
- "Model": "yolo_v8n",
+ "Platform": "Intel® Xeon® Gold 5218T CPU-only",
+ "Model": "ssd-resnet34-1200",
"featured_SKU": false,
"whats_new_model": false,
"PlatformType": "Intel® Xeon®, CPU-only",
@@ -11070,9 +10930,9 @@
"Precisions": [
{
"int4": "",
- "int8": "",
+ "int8": 17.6,
"fp16": "",
- "fp32": "",
+ "fp32": 4.58,
"bf16": ""
}
],
@@ -11083,7 +10943,7 @@
"Precisions": [
{
"int4": "",
- "int8": 3.9,
+ "int8": 116.38,
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11096,18 +10956,18 @@
},
{
"Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "bert-base-cased",
+ "Model": "yolo11",
"featured_SKU": false,
- "whats_new_model": false,
+ "whats_new_model": true,
"PlatformType": "Intel® Xeon®, CPU-only",
"Parameters": {
"throughput": {
"Precisions": [
{
"int4": "",
- "int8": 216.65,
+ "int8": "",
"fp16": "",
- "fp32": 80.03,
+ "fp32": 206.34,
"bf16": ""
}
],
@@ -11118,7 +10978,7 @@
"Precisions": [
{
"int4": "",
- "int8": 14.43,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11131,7 +10991,7 @@
},
{
"Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "mask_rcnn_resnet50_atrous_coco",
+ "Model": "yolo_v8n",
"featured_SKU": false,
"whats_new_model": false,
"PlatformType": "Intel® Xeon®, CPU-only",
@@ -11140,9 +11000,9 @@
"Precisions": [
{
"int4": "",
- "int8": 3.15,
+ "int8": 451.74,
"fp16": "",
- "fp32": 0.91,
+ "fp32": 174.3,
"bf16": ""
}
],
@@ -11153,7 +11013,7 @@
"Precisions": [
{
"int4": "",
- "int8": "",
+ "int8": 5.96,
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11165,8 +11025,8 @@
}
},
{
- "Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "mobilenet-v2",
+ "Platform": "Intel® Xeon® Gold 6338N CPU-only",
+ "Model": "bert-base-cased",
"featured_SKU": false,
"whats_new_model": false,
"PlatformType": "Intel® Xeon®, CPU-only",
@@ -11175,9 +11035,9 @@
"Precisions": [
{
"int4": "",
- "int8": 5447.03,
+ "int8": 622.61,
"fp16": "",
- "fp32": 1910.7,
+ "fp32": 241.02,
"bf16": ""
}
],
@@ -11188,182 +11048,7 @@
"Precisions": [
{
"int4": "",
- "int8": 1.38,
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "resnet-50",
- "featured_SKU": false,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": 958.08,
- "fp16": "",
- "fp32": 269.68,
- "bf16": ""
- }
- ],
- "Unit": "FPS",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 3.08,
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "ssd-resnet34-1200",
- "featured_SKU": false,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": 17.6,
- "fp16": "",
- "fp32": 4.58,
- "bf16": ""
- }
- ],
- "Unit": "FPS",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 116.38,
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "yolo11",
- "featured_SKU": false,
- "whats_new_model": true,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": 206.34,
- "bf16": ""
- }
- ],
- "Unit": "FPS",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Xeon® Gold 5218T CPU-only",
- "Model": "yolo_v8n",
- "featured_SKU": false,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": 451.74,
- "fp16": "",
- "fp32": 174.3,
- "bf16": ""
- }
- ],
- "Unit": "FPS",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 5.96,
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Xeon® Gold 6338N CPU-only",
- "Model": "bert-base-cased",
- "featured_SKU": false,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": 622.61,
- "fp16": "",
- "fp32": 241.02,
- "bf16": ""
- }
- ],
- "Unit": "FPS",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 6.35,
+ "int8": 6.35,
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11387,7 +11072,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 2.45,
+ "fp32": "",
"bf16": ""
}
],
@@ -11632,7 +11317,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 2.25,
+ "fp32": "",
"bf16": ""
}
],
@@ -11875,7 +11560,7 @@
"Precisions": [
{
"int4": "",
- "int8": 0.18,
+ "int8": "",
"fp16": "",
"fp32": "",
"bf16": ""
@@ -11888,13 +11573,13 @@
"Precisions": [
{
"int4": "",
- "int8": 5505.78,
+ "int8": 113.3,
"fp16": "",
"fp32": "",
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -12087,7 +11772,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 3.44,
+ "fp32": "",
"bf16": ""
}
],
@@ -12384,42 +12069,7 @@
"bf16": ""
}
],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
- {
- "Platform": "Intel® Xeon® Platinum 8380 CPU-only",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": false,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 12.65,
- "fp16": 12.92,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -12540,8 +12190,8 @@
"Precisions": [
{
"int4": "",
- "int8": 0.73,
- "fp16": 0.64,
+ "int8": "",
+ "fp16": "",
"fp32": "",
"bf16": ""
}
@@ -12553,13 +12203,13 @@
"Precisions": [
{
"int4": "",
- "int8": 1357.53,
- "fp16": 1544.72,
+ "int8": 27.6,
+ "fp16": "",
"fp32": "",
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -12752,7 +12402,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 5.18,
+ "fp32": "",
"bf16": ""
}
],
@@ -13019,41 +12669,6 @@
}
}
},
- {
- "Platform": "Intel® Xeon® Platinum 8480+ CPU-only",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": true,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 3.36,
- "fp16": 3.31,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
{
"Platform": "Intel® Xeon® Platinum 8480+ CPU-only",
"Model": "stable-diffusion-v1-5",
@@ -13084,7 +12699,7 @@
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -13205,8 +12820,8 @@
"Precisions": [
{
"int4": "",
- "int8": 0.77,
- "fp16": 0.68,
+ "int8": "",
+ "fp16": "",
"fp32": "",
"bf16": ""
}
@@ -13218,13 +12833,13 @@
"Precisions": [
{
"int4": "",
- "int8": 1290.6,
- "fp16": 1468.28,
+ "int8": 26.2,
+ "fp16": "",
"fp32": "",
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -13417,7 +13032,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 6.41,
+ "fp32": "",
"bf16": ""
}
],
@@ -13684,41 +13299,6 @@
}
}
},
- {
- "Platform": "Intel® Xeon® Platinum 8580 CPU-only",
- "Model": "stable-diffusion-v1-5",
- "featured_SKU": true,
- "whats_new_model": false,
- "PlatformType": "Intel® Xeon®, CPU-only",
- "Parameters": {
- "throughput": {
- "Precisions": [
- {
- "int4": "",
- "int8": "",
- "fp16": "",
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "Tokens per sec",
- "UnitDesc": "higher is better"
- },
- "latency": {
- "Precisions": [
- {
- "int4": "",
- "int8": 3.03,
- "fp16": 2.94,
- "fp32": "",
- "bf16": ""
- }
- ],
- "Unit": "ms",
- "UnitDesc": "lower is better"
- }
- }
- },
{
"Platform": "Intel® Xeon® Platinum 8580 CPU-only",
"Model": "stable-diffusion-v1-5",
@@ -13749,7 +13329,7 @@
"bf16": ""
}
],
- "Unit": "ms",
+ "Unit": "sec",
"UnitDesc": "lower is better"
}
}
@@ -13874,7 +13454,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.22,
+ "fp32": "",
"bf16": ""
}
],
@@ -14080,7 +13660,7 @@
"Platform": "Intel® Core™ Ultra 7 Processor 265H",
"Model": "flux.1-schnell",
"featured_SKU": true,
- "whats_new_model": false,
+ "whats_new_model": true,
"PlatformType": "Intel® Core™, iGPU-only",
"IE_type": "iGPU",
"Parameters": {
@@ -14123,7 +13703,7 @@
"throughput": {
"Precisions": [
{
- "int4": 45.3,
+ "int4": "",
"int8": "",
"fp16": "",
"fp32": "",
@@ -14136,7 +13716,7 @@
"latency": {
"Precisions": [
{
- "int4": 22.05,
+ "int4": "",
"int8": "",
"fp16": "",
"fp32": "",
@@ -14159,7 +13739,7 @@
"throughput": {
"Precisions": [
{
- "int4": 64.5,
+ "int4": 12.6,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14172,7 +13752,7 @@
"latency": {
"Precisions": [
{
- "int4": 15.5,
+ "int4": 79.5,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14195,7 +13775,7 @@
"throughput": {
"Precisions": [
{
- "int4": 69.7,
+ "int4": 13.1,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14208,7 +13788,7 @@
"latency": {
"Precisions": [
{
- "int4": 14.3,
+ "int4": 76.1,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14231,7 +13811,7 @@
"throughput": {
"Precisions": [
{
- "int4": 72.1,
+ "int4": 12.5,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14244,7 +13824,7 @@
"latency": {
"Precisions": [
{
- "int4": 13.9,
+ "int4": 79.8,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14267,7 +13847,7 @@
"throughput": {
"Precisions": [
{
- "int4": 121.4,
+ "int4": 27.2,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14280,7 +13860,7 @@
"latency": {
"Precisions": [
{
- "int4": 8.2,
+ "int4": 36.7,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14305,7 +13885,7 @@
{
"int4": "",
"int8": "",
- "fp16": 1.96,
+ "fp16": "",
"fp32": "",
"bf16": ""
}
@@ -14339,7 +13919,7 @@
"throughput": {
"Precisions": [
{
- "int4": 70.3,
+ "int4": 14.0,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14352,7 +13932,7 @@
"latency": {
"Precisions": [
{
- "int4": 14.2,
+ "int4": 71.6,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14411,7 +13991,7 @@
"throughput": {
"Precisions": [
{
- "int4": 108.5,
+ "int4": 19.9,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14424,7 +14004,7 @@
"latency": {
"Precisions": [
{
- "int4": 9.2,
+ "int4": 50.2,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14447,7 +14027,7 @@
"throughput": {
"Precisions": [
{
- "int4": 78.4,
+ "int4": 13.3,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14460,7 +14040,7 @@
"latency": {
"Precisions": [
{
- "int4": 12.8,
+ "int4": 75.0,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14483,7 +14063,7 @@
"throughput": {
"Precisions": [
{
- "int4": 78.4,
+ "int4": 13.3,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14496,7 +14076,7 @@
"latency": {
"Precisions": [
{
- "int4": 12.8,
+ "int4": 75.2,
"int8": "",
"fp16": "",
"fp32": "",
@@ -14605,8 +14185,8 @@
"Precisions": [
{
"int4": "",
- "int8": 5.25,
- "fp16": 5.1,
+ "int8": 5.4,
+ "fp16": 5.4,
"fp32": "",
"bf16": ""
}
@@ -14688,6 +14268,42 @@
}
}
},
+ {
+ "Platform": "Intel® Core™ Ultra 7 Processor 265H",
+ "Model": "llama-2-7b-chat",
+ "featured_SKU": true,
+ "whats_new_model": true,
+ "PlatformType": "Intel® Core™, NPU-only",
+ "IE_type": "NPU",
+ "Parameters": {
+ "throughput": {
+ "Precisions": [
+ {
+ "int4": 10.4,
+ "int8": "",
+ "fp16": "",
+ "fp32": "",
+ "bf16": ""
+ }
+ ],
+ "Unit": "Tokens per sec",
+ "UnitDesc": "higher is better"
+ },
+ "latency": {
+ "Precisions": [
+ {
+ "int4": 95.8,
+ "int8": "",
+ "fp16": "",
+ "fp32": "",
+ "bf16": ""
+ }
+ ],
+ "Unit": "ms",
+ "UnitDesc": "lower is better"
+ }
+ }
+ },
{
"Platform": "Intel® Core™ Ultra 7 Processor 265H",
"Model": "mask_rcnn_resnet50_atrous_coco",
@@ -14724,6 +14340,78 @@
}
}
},
+ {
+ "Platform": "Intel® Core™ Ultra 7 Processor 265H",
+ "Model": "mistral-7B-instruct-V0.2",
+ "featured_SKU": true,
+ "whats_new_model": true,
+ "PlatformType": "Intel® Core™, NPU-only",
+ "IE_type": "NPU",
+ "Parameters": {
+ "throughput": {
+ "Precisions": [
+ {
+ "int4": 8.0,
+ "int8": "",
+ "fp16": "",
+ "fp32": "",
+ "bf16": ""
+ }
+ ],
+ "Unit": "Tokens per sec",
+ "UnitDesc": "higher is better"
+ },
+ "latency": {
+ "Precisions": [
+ {
+ "int4": 125.6,
+ "int8": "",
+ "fp16": "",
+ "fp32": "",
+ "bf16": ""
+ }
+ ],
+ "Unit": "ms",
+ "UnitDesc": "lower is better"
+ }
+ }
+ },
+ {
+ "Platform": "Intel® Core™ Ultra 7 Processor 265H",
+ "Model": "phi-3-mini-4k-instruct",
+ "featured_SKU": true,
+ "whats_new_model": true,
+ "PlatformType": "Intel® Core™, NPU-only",
+ "IE_type": "NPU",
+ "Parameters": {
+ "throughput": {
+ "Precisions": [
+ {
+ "int4": 11.1,
+ "int8": "",
+ "fp16": "",
+ "fp32": "",
+ "bf16": ""
+ }
+ ],
+ "Unit": "Tokens per sec",
+ "UnitDesc": "higher is better"
+ },
+ "latency": {
+ "Precisions": [
+ {
+ "int4": 90.1,
+ "int8": "",
+ "fp16": "",
+ "fp32": "",
+ "bf16": ""
+ }
+ ],
+ "Unit": "ms",
+ "UnitDesc": "lower is better"
+ }
+ }
+ },
{
"Platform": "Intel® Core™ Ultra 7 Processor 265H",
"Model": "mobilenet-v2",
@@ -14918,7 +14606,7 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 0.24,
+ "fp32": "",
"bf16": ""
}
],
@@ -15314,8 +15002,8 @@
"int4": "",
"int8": "",
"fp16": "",
- "fp32": 11.92,
- "bf16": 83.42
+ "fp32": "",
+ "bf16": ""
}
],
"Unit": "FPS",