model in TVM ● Support Any-dim in typing ● Use shape function to compute the type at runtime ● Virtual machine as a new runtime for Relay ● Dynamic codegen (WIP) ○ Kernel dispatch for a single op ○ Data dependent© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Relay virtual machine Relay Executable relay.vm.compile Relay Object (hardware independent) Code segment VM a data type using the entries from a register. AllocClosure Allocates a closure with a lowered virtual machine function. If Jumps to the true or false offset depending on the condition. Goto Unconditionally

areas: o Core Infrastructure Improvements to TVM o_uTVM: support for microcontrollers in TVM o_ Virtual Machine and dynamic NNs support (w/ AWS folks) o_ Improved NLP support, with focus on transformers o QQ octoML BERT has many reshape operations, which are currently implemented using copy, 10 Virtual Machine e Many improvements from contributors at UW, AWS, and OctoML. e Initial implementation

Erica Virtual Assistant (6/18) Note: We assume a start at zero users from Erica’s launch in 6/18. Pilot users excluded. Source: Bank of America (2/21, 4/24, 2/25) Bank of America Erica Virtual Assistant 10/21 3/22 8/22 1/23 6/23 11/23 4/24 9/24 2/25 Cumulative Client Interactions with Erica Virtual Assistant (MM) Note: Erica is a conversational AI built into Bank of America’s mobile app that helps

cases where close physical proximity of the team is not practical or feasible, programs can use virtual teams with synchronized daily meetings to develop predictable and routine meeting schedules to enhance fund travel by stakeholders across the user community; alternatively, or in addition, provide for virtual participation.  Arrange for developers to demonstrate existing capabilities, prototypes, and

remote_mod = remote.load_module(“mylib.so") func = remote_mod[“npufunction0"] func(remote_a, remote_b)Virtual Machine: Supporting Dynamic Workload Dynamic shape workloads More runtime objects: Arrays, Tuples

buffer to hide memory load latency 。 storage align to reduce bank conflicts of shared memory 。 Virtual threads for data reuse (on going) Performance on V100 (FP16) 计算平台事业部 COMPUTING PLATFORM 512, 16

static dispatch, i.e. you need to annotate a function as virtual, in order to have dynamic dispatch. On the other hand, in Julia every method is "virtual" (although it's more general than that since methods The default is 512 MB. • old_prefix: the virtual file directory in which to place old files on one side of the diff. Default is "a". • new_prefix: the virtual file directory in which to place new files recursion_limit: the upper limit on the number of merges of common ancestors to perform to try to build a new virtual merge base for the merge. The default is no limit. This field is only present on libgit2 versions

static dispatch, i.e. you need to annotate a function as virtual, in order to have dynamic dispatch. On the other hand, in Julia every method is "virtual" (although it's more general than that since methods The default is 512 MB. • old_prefix: the virtual file directory in which to place old files on one side of the diff. Default is "a". • new_prefix: the virtual file directory in which to place new files recursion_limit: the upper limit on the number of merges of common ancestors to perform to try to build a new virtual merge base for the merge. The default is no limit. This field is only present on libgit2 versions

Product Note: Operator Operator is an example of OpenAI’s agentic approach. Leveraging its own virtual browser, Operator can navigate the web, click on buttons, fill in forms, and gather data just like

static dispatch, i.e. you need to annotate a function as virtual, in order to have dynamic dispatch. On the other hand, in Julia every method is "virtual" (although it's more general than that since methods The default is 512 MB. • old_prefix: the virtual file directory in which to place old files on one side of the diff. Default is "a". • new_prefix: the virtual file directory in which to place new files recursion_limit: the upper limit on the number of merges of common ancestors to perform to try to build a new virtual merge base for the merge. The default is no limit. This field is only present on libgit2 versions