Bug fixed: the device_group arg is deprecated from compile and added in generate method,

smedhe · smedhe · commit a8661cd751bc · 2025-08-05T10:44:44.000Z
changes made in jupyter notebooks

Signed-off-by: Sharvari Medhe &lt;smedhe@qti.qualcomm.com&gt;
diff --git a/notebooks/QEfficientGPT2.ipynb b/notebooks/QEfficientGPT2.ipynb
@@ -94,9 +94,10 @@
     "\n",
     "qeff_model.compile(\n",
     "    num_cores=14,\n",
-    "    mxfp6=True,\n",
-    "    device_group=[0],\n",
-    ")"
+    "    mxfp6=True\n",
+    ")\n",
+    "\n",
+    "#the device_group arg is deprecated from compile and added in generate method"
    ]
   },
   {
@@ -117,7 +118,7 @@
     "# post compilation, we can print the latency stats for the kv models, We provide API to print token and Latency stats on Cloud AI 100\n",
     "# We need the compiled prefill and decode qpc to compute the token generated, This is based on Greedy Sampling Approach\n",
     "\n",
-    "qeff_model.generate(prompts=[\"My name is\"])"
+    "qeff_model.generate(prompts=[\"My name is\"], device_group=[0])"
    ]
   }
  ],
diff --git a/notebooks/QEfficientMPT.ipynb b/notebooks/QEfficientMPT.ipynb
@@ -93,9 +93,9 @@
     "\n",
     "qeff_model.compile(\n",
     "    num_cores=14,\n",
-    "    mxfp6=True,\n",
-    "    device_group=[0],\n",
-    ")"
+    "    mxfp6=True\n",
+    ")\n",
+    "# the device_group arg is deprecated from compile and added in generate method"
    ]
   },
   {
@@ -116,7 +116,7 @@
     "# post compilation, we can print the latency stats for the kv models, We provide API to print token and Latency stats on Cloud AI 100\n",
     "# We need the compiled prefill and decode qpc to compute the token generated, This is based on Greedy Sampling Approach\n",
     "\n",
-    "qeff_model.generate(prompts=[\"My name is\"])"
+    "qeff_model.generate(prompts=[\"My name is\"], device_group=[0])"
    ]
   }
  ],

Original file line number	Diff line number	Diff line change
`@@ -94,9 +94,10 @@`
`94`	`94`	`"\n",`
`95`	`95`	`"qeff_model.compile(\n",`
`96`	`96`	`" num_cores=14,\n",`
`97`		`- " mxfp6=True,\n",`
`98`		`- " device_group=[0],\n",`
`99`		`- ")"`
	`97`	`+ " mxfp6=True\n",`
	`98`	`+ ")\n",`
	`99`	`+ "\n",`
	`100`	`+ "#the device_group arg is deprecated from compile and added in generate method"`
`100`	`101`	`]`
`101`	`102`	`},`
`102`	`103`	`{`
`@@ -117,7 +118,7 @@`
`117`	`118`	`"# post compilation, we can print the latency stats for the kv models, We provide API to print token and Latency stats on Cloud AI 100\n",`
`118`	`119`	`"# We need the compiled prefill and decode qpc to compute the token generated, This is based on Greedy Sampling Approach\n",`
`119`	`120`	`"\n",`
`120`		`- "qeff_model.generate(prompts=[\"My name is\"])"`
	`121`	`+ "qeff_model.generate(prompts=[\"My name is\"], device_group=[0])"`
`121`	`122`	`]`
`122`	`123`	`}`
`123`	`124`	`],`
Original file line number	Diff line number	Diff line change
`@@ -93,9 +93,9 @@`
`93`	`93`	`"\n",`
`94`	`94`	`"qeff_model.compile(\n",`
`95`	`95`	`" num_cores=14,\n",`
`96`		`- " mxfp6=True,\n",`
`97`		`- " device_group=[0],\n",`
`98`		`- ")"`
	`96`	`+ " mxfp6=True\n",`
	`97`	`+ ")\n",`
	`98`	`+ "# the device_group arg is deprecated from compile and added in generate method"`
`99`	`99`	`]`
`100`	`100`	`},`
`101`	`101`	`{`
`@@ -116,7 +116,7 @@`
`116`	`116`	`"# post compilation, we can print the latency stats for the kv models, We provide API to print token and Latency stats on Cloud AI 100\n",`
`117`	`117`	`"# We need the compiled prefill and decode qpc to compute the token generated, This is based on Greedy Sampling Approach\n",`
`118`	`118`	`"\n",`
`119`		`- "qeff_model.generate(prompts=[\"My name is\"])"`
	`119`	`+ "qeff_model.generate(prompts=[\"My name is\"], device_group=[0])"`
`120`	`120`	`]`
`121`	`121`	`}`
`122`	`122`	`],`