Skip to content

Agent FFI integration #119

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 10 commits into
base: main
Choose a base branch
from
+3,434 −0
Open

Agent FFI integration #119

Show file tree
Hide file tree
Changes from 7 commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
3e2e893
feat: add FFI prompt tests with environment-based API keys
zanderjiang Nov 10, 2025
25092ad
formatting
zanderjiang Nov 10, 2025
365536e
full ffi prompt
zanderjiang Nov 17, 2025
ac2ee6f
agent vibecoding end-to-end example
zanderjiang Nov 17, 2025
46ca5a5
Merge branch 'flashinfer-ai:main' into main
zanderjiang Nov 17, 2025
2f5af2f
reformat tests
zanderjiang Nov 17, 2025
b7d62c3
Merge branch 'main' of https://github.com/zanderjiang/flashinfer-benc…
zanderjiang Nov 17, 2025
5d60fe9
Merge branch 'flashinfer-ai:main' into main
zanderjiang Nov 20, 2025
9a0ce21
remove testing results
zanderjiang Nov 26, 2025
19a32bb
updated agent vibecode ffi example
zanderjiang Nov 28, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,3 +47,5 @@ web/**/public

tmp/
AGENTS.md

.env
48 changes: 48 additions & 0 deletions examples/ffi/Example-FlashInfer-Trace/definitions/gemm_n4096_k4096.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,48 @@
{
"name": "gemm_n4096_k4096",
"description": "General matrix multiply (GEMM) C = A @ B.T. Captured from Llama 3.1 8B attn.o_proj.",
"op_type": "gemm",
"tags": [
"status:verified",
"model:llama-3.1-8b"
],
"axes": {
"M": {
"type": "var"
},
"N": {
"type": "const",
"value": 4096
},
"K": {
"type": "const",
"value": 4096
}
},
"inputs": {
"A": {
"shape": [
"M",
"K"
],
"dtype": "float16"
},
"B": {
"shape": [
"N",
"K"
],
"dtype": "float16"
}
},
"outputs": {
"C": {
"shape": [
"M",
"N"
],
"dtype": "float16"
}
},
"reference": "import torch\n\ndef run(A, B):\n C = torch.matmul(A, B.T)\n return C"
}
Empty file added examples/ffi/Example-FlashInfer-Trace/solutions/agent_vibecode_gemm.json
View file
Open in desktop
Empty file.
43 changes: 43 additions & 0 deletions examples/ffi/Example-FlashInfer-Trace/workloads/gemm_n4096_k4096.jsonl
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,43 @@
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "280860e6-08f0-427c-b7c5-9cffcfab1a10", "axes": {"M": 256}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "2e90109a-282e-484f-b94b-61f49e72fde2", "axes": {"M": 248}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "e7c939ae-2083-4b6f-a51a-8c76ffd08926", "axes": {"M": 240}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "3ab6479f-c1d0-4743-b2d4-1a46b01f9db7", "axes": {"M": 232}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "29581017-6470-4d78-9d02-554adecd9822", "axes": {"M": 224}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "e97cc8b3-9a2f-4d0d-aaa9-7522413c78da", "axes": {"M": 216}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "c69f1da6-7c62-46f5-867e-3cf5ed3aac04", "axes": {"M": 208}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "38a98eb2-3a86-41ce-994c-b6a2cec932a6", "axes": {"M": 200}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "62f7844c-b1f5-4e08-a057-70e55f092931", "axes": {"M": 192}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "1342c570-b505-4ccf-9fc2-b377d25b397e", "axes": {"M": 184}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "beb5da20-954c-44ef-8f1b-35a5dd848b06", "axes": {"M": 176}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "c8c9178e-65f2-4124-a322-e66987aa1b34", "axes": {"M": 168}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "c103a7db-34d0-4bc0-abb0-2833c2458c50", "axes": {"M": 160}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "ed0bbc00-57e9-46e0-af02-249a64a46fa0", "axes": {"M": 152}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "83deae78-6557-46e9-b3ef-2ed254192d13", "axes": {"M": 144}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "76fde2b6-3f6e-484f-bcd4-fca79272a690", "axes": {"M": 136}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "29ebd771-0f1c-4894-8532-7265275a02b1", "axes": {"M": 128}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "73dd121d-72c7-4dda-9b6e-37d55c6ed867", "axes": {"M": 120}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "27fa4c34-6e08-459a-a8c4-f37b7cdb037b", "axes": {"M": 112}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "faac134d-36e3-4f99-809c-9e544ea5216f", "axes": {"M": 104}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "1948bb55-6253-4b9c-aa17-4fb13ffef7d0", "axes": {"M": 96}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "3a7e6db4-4127-45c1-9e54-c8c7cc25d632", "axes": {"M": 88}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "dcb04a7e-2faa-4858-a495-71c658b299ad", "axes": {"M": 80}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "ef996a93-b3b6-4702-aae2-f28fdbcfdd48", "axes": {"M": 72}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "67d4c8f3-2ff5-4838-8f07-d5d16f602eb3", "axes": {"M": 64}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "54d34708-309d-462b-829a-74c90243093c", "axes": {"M": 56}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "59ca23f5-a523-4cc4-9c1f-db510753d3f4", "axes": {"M": 48}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "54062a8b-a9ca-47d8-b5ef-7f6f0325ef39", "axes": {"M": 40}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "5230e6ed-48b8-4765-bc9f-a7cdaabed615", "axes": {"M": 32}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "73212638-6584-476b-848d-2cb8ce0b829c", "axes": {"M": 24}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "404d21b1-2237-4e3c-b3ff-9b68878e5d70", "axes": {"M": 16}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "4cdab8cd-cb6b-4e73-8fe0-75b55fd784b5", "axes": {"M": 8}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "f439da26-2483-406c-977b-be185901207f", "axes": {"M": 4}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "6c2f4ba8-94d3-4e8f-997f-b7454242695a", "axes": {"M": 2}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "e39649a6-6f42-4a1b-9731-b45a9a87f7a5", "axes": {"M": 1}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "094ef833-829f-4efa-925d-d5bae9d6a116", "axes": {"M": 7}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "4c6bdefd-dd94-48b6-be3e-7eb25658eefd", "axes": {"M": 35}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "da2a2234-f5e9-4332-b62d-39865128153c", "axes": {"M": 972}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "897b6544-56c2-4d96-98fd-453ae3418e4b", "axes": {"M": 70}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "8a8311fa-8bb6-487d-8a36-7378e9680df8", "axes": {"M": 2053}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "b626104c-94cc-436b-9d2c-1d31432c1a87", "axes": {"M": 8192}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "547e0ce5-e484-4e0c-8b38-f153fb7ce6d4", "axes": {"M": 2379}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
{"definition": "gemm_n4096_k4096", "solution": null, "workload": {"uuid": "339de815-896d-4d4c-8060-07208d559276", "axes": {"M": 15}, "inputs": {"A": {"type": "random"}, "B": {"type": "random"}}}, "evaluation": null}
Loading
Loading