Comparing
llm-generate-upgrade
(7c6e18f
) withdevelop
(e830e25
)+2%
- Improvements
- 0
- Regressions
- 0
- Untouched
- 1
- New
- 0
- Dropped
- 0
- Ignored
- 0
Benchmarks
Passed
test_cache_time
tests/integration/test_cache.py::test_cache_time+2%
551 ms
540.7 ms
Commits
Click on a commit to change the comparison rangeBase
develop
e830e25
0%
Refactor llm generation to return generations and statistics
dbcfafa
10 months ago
by plaguss+1%
First version of async llms with statistics
394984f
10 months ago
by plaguss0%
Return generations with list of strings and token count from _raw_response
9003fa0
10 months ago
by plaguss0%
Passing tests for inference endpoints
4cf0e2f
10 months ago
by plaguss-2%
Testing vLLM with statistics
1b6d15c
10 months ago
by plaguss0%
Refactor to remove code duplication
608e8b6
10 months ago
by plaguss+1%
Fix tests from merge responses and group generations
c8063a4
10 months ago
by plaguss0%
Move import to guarded type hint
6f6769a
10 months ago
by plaguss-2%
Update tests with the llm generate output format
241d899
10 months ago
by plaguss0%
Merge branch 'develop' of https://github.com/argilla-io/distilabel into llm-generate-upgrade
a657859
10 months ago
by plaguss+2%
Fix pending tests
40e408d
10 months ago
by plaguss0%
Fix test failing with vllm version upgrade
6c2e1fd
10 months ago
by plaguss0%
Another fix including tokenizer for our llm to work and to avoid outlines complaining
d28a798
10 months ago
by plaguss0%
Fix dummy offline batch generation
1bc28ba
10 months ago
by plaguss+1%
Merge and fix conflict
13f42b0
10 months ago
by plaguss-1%
Compute tokens using the tokenizer if available
edbea28
10 months ago
by plaguss0%
Update docs to include references to the new outputs of the LLMs including statistics
e97f901
10 months ago
by plaguss+1%
Fix count_tokens and update corresponding tests
98eb9a2
9 months ago
by plaguss0%
Make a single list comprehension as per code review
7c6e18f
9 months ago
by plaguss