root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-24 10:24:35 +00:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

server : add tests
Some checks failed
Python Type-Check / pyright type-check (push) Has been cancelled
Details

Browse Source
This commit is contained in:
Georgi Gerganov 2024-12-04 13:11:26 +02:00
parent b436edaad9
commit 81611bef72
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 31 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
examples/server/tests/unit/test_speculative.py
View File

@ -82,6 +82,37 @@ def test_different_draft_min_draft_max():
last_content = res.body["content"]
def test_slot_ctx_not_exceeded():
global server
server.n_ctx = 64
server.start()
res = server.make_request("POST", "/completion", data={
"prompt": "Hello " * 56,
"temperature": 0.0,
"top_k": 1,
"speculative.p_min": 0.0,
})
assert res.status_code == 200
assert len(res.body["content"]) > 0
def test_with_ctx_shift():
global server
server.n_ctx = 64
server.start()
res = server.make_request("POST", "/completion", data={
"prompt": "Hello " * 56,
"temperature": 0.0,
"top_k": 1,
"n_predict": 64,
"speculative.p_min": 0.0,
})
assert res.status_code == 200
assert len(res.body["content"]) > 0
assert res.body["tokens_predicted"] == 64
assert res.body["truncated"] == True
@pytest.mark.parametrize("n_slots,n_requests", [
(1, 2),
(2, 2),