MOVZX
1 Followers · 0 Following
Repositories Projects Packages Public Activity Starred Repositories
MOVZX pushed to main at iHore/vLLM-4x-RTX-3090 2026-10-15 13:24:49 +07:00
eb8c302bb3 fix: Adjust start script to avoid crashing on high pressure KV cache
MOVZX pushed to main at iHore/vLLM-4x-RTX-3090 2026-10-15 13:23:54 +07:00
771e4a4c57 fix: Adjust start script to avoid crashing on high pressure KV cache
MOVZX pushed to main at iHore/vLLM-4x-RTX-3090 2026-10-15 13:19:49 +07:00
adc0e941fc fix: Adjust start script to avoid crashing on high pressure KV cache
MOVZX pushed to main at iHore/vLLM-4x-RTX-3090 2026-10-15 13:19:26 +07:00
b3a040ee98 chore: Adjust start script to avoid crashing on high pressure KV cache
MOVZX pushed to main at iHore/vLLM-4x-RTX-3090 2026-10-15 01:03:09 +07:00
777e910dd0 Initial commit
MOVZX created branch main in iHore/vLLM-4x-RTX-3090 2026-10-15 01:03:09 +07:00
MOVZX created repository iHore/vLLM-4x-RTX-3090 2026-10-15 01:02:35 +07:00
MOVZX pushed to patches at iHore/vLLM 2026-10-14 15:12:23 +07:00
e7cf27cc6f [Model] Fix runtime abliteration for Qwen3.5 models and Qwen3-Next MTP
MOVZX pushed to patches at iHore/vLLM 2026-10-14 14:58:45 +07:00
8151c7b158 [Model] Add opt-in runtime abliteration for Qwen3.5/Qwen3.8
MOVZX pushed to patches at iHore/vLLM 2026-10-14 12:22:49 +07:00
ac305c44cb Merge branch 'main' into patches
89ac8d971b [CI][Test] Allow 10% ngram acceptance drop under preemption on CUDA in test_async_scheduling_accuracy (#60715)
b74233860d [CI][ROCm] Wait for VRAM to settle before building LLM in test_full_graph (#60872)
75e3b17c05 [Frontend] Add cache salt support to generative scoring (#60841)
90ba2f34b5 [Build] Find compatible published wheels for precompiled installs (#58871)
Compare 316 commits »
MOVZX pushed to patches at iHore/vLLM 2026-10-10 00:02:27 +07:00
MOVZX pushed to main at iHore/vLLM 2026-10-10 00:01:16 +07:00
89ac8d971b [CI][Test] Allow 10% ngram acceptance drop under preemption on CUDA in test_async_scheduling_accuracy (#60715)
b74233860d [CI][ROCm] Wait for VRAM to settle before building LLM in test_full_graph (#60872)
75e3b17c05 [Frontend] Add cache salt support to generative scoring (#60841)
90ba2f34b5 [Build] Find compatible published wheels for precompiled installs (#58871)
970dc63478 [Docs] Add Giga-Embeddings bidirectional Qwen3 usage to embed docs (#60854)
Compare 315 commits »
MOVZX pushed to patches at iHore/vLLM 2026-10-09 21:37:55 +07:00
31bee2c2fd [Qwen4Exp] Repair: import Qwen4ExpTextConfig from transformers
bf3e4af6b1 [KV Cache] Hold a finished turn's prefix until its next turn is admitted
f5d9021745 [Scheduler] Release a waiter's pin on the deferred free path too
633c758d4c [Scheduler] Pin a waiting request's prefix-cache hit until it is admitted
95ba99ed29 [Mamba] Key the prompt tail when it lands on a block end
Compare 79 commits »
MOVZX pushed to main at iHore/SWE-bench_Verified 2026-07-17 09:03:38 +07:00
91aa3ed51b Update eval.yaml (#4)
880e32832a Add eval.yaml (#2)
99450355ca Update README.md
fd80552a1f Upload dataset
2d98c78b4c initial commit
MOVZX created branch main in iHore/SWE-bench_Verified 2026-07-17 09:03:38 +07:00
MOVZX created repository iHore/SWE-bench_Verified 2026-07-17 09:03:28 +07:00
MOVZX created branch main in iHore/SWE-bench_Pro 2026-07-17 09:02:03 +07:00
MOVZX pushed to main at iHore/SWE-bench_Pro 2026-07-17 09:02:03 +07:00
7ab5114912 Add eval.yaml (#4)
2dd05cab15 Upload dataset
feb7159156 Remove old CSV file: test.csv
b662601616 Remove old CSV file: external_hf_v2.csv
f60c8c4f21 Remove old CSV file: external_hf.csv
Compare 10 commits »
MOVZX created repository iHore/SWE-bench_Pro 2026-07-17 09:00:21 +07:00