Commit Graph

Select branches

Hide Pull Requests

main

Mono Color

• a7a2c722af fix large completion probe timeout main the magician 2026-07-15 22:20:25 +08:00
• 163da44dfd record current performance experiments the magician 2026-07-15 21:11:32 +08:00
• 9fc7e98af4 fastpath oversized basic test outputs the magician 2026-07-15 17:18:39 +08:00
• 2b7880efa7 fix platform basic test compatibility the magician 2026-07-15 14:25:52 +08:00
• e374b14bf9 add moe bucket profiling and record bottlenecks the magician 2026-07-15 11:34:59 +08:00
• 8b4ee2ff8b record max batched tokens 8192 experiment the magician 2026-07-15 10:44:24 +08:00
• 2c7a8d12b3 record max batched tokens 32768 experiment the magician 2026-07-15 10:27:33 +08:00
• 59e497fdc9 disable profiling trace by default the magician 2026-07-15 04:43:46 +08:00
• cb962589e9 revert moe router topk softmax experiment the magician 2026-07-15 04:21:03 +08:00
• eea0653ace experiment moe router topk softmax the magician 2026-07-15 04:19:56 +08:00
• 16d52b469c record 245k platform-equivalent benchmark the magician 2026-07-15 03:55:20 +08:00
• 3cf62afaaa stabilize platform launch and record official-like baseline the magician 2026-07-15 03:23:39 +08:00
• f49b8a51ec document docker build line ending fix the magician 2026-07-15 01:54:56 +08:00
• f3303b3b26 enforce lf line endings for docker build scripts the magician 2026-07-15 01:40:40 +08:00
• 4b89576cbf reduce docker build context and match working baseline the magician 2026-07-15 01:32:21 +08:00
• d1f8048b6e record moe tiny batch matrix results the magician 2026-07-15 01:28:06 +08:00
• af375830c7 align platform build config with working submission the magician 2026-07-15 00:49:24 +08:00
• 62f10c3375 restore official docker build baseline the magician 2026-07-14 23:03:24 +08:00
• 3a06e4f0e5 restore standard Dockerfile submission name the magician 2026-07-14 22:37:08 +08:00
• 3ec5e7911f normalize DockerFile formatting the magician 2026-07-14 22:19:00 +08:00
• ae64b20e22 match official DockerFile submission name the magician 2026-07-14 22:17:12 +08:00
• 2c6cabcc63 optimize moe tiny batch tokenwise path the magician 2026-07-14 21:49:55 +08:00
• 0597fa6c6d profile decode path and optimize tiny batch moe the magician 2026-07-14 20:44:05 +08:00
• a5abd66e21 test eager and custom allreduce toggles the magician 2026-07-14 19:18:46 +08:00
• 82c1463b0c document multi-step scheduler diagnostics the magician 2026-07-14 18:17:42 +08:00
• 465a9c8995 add decode throughput analysis the magician 2026-07-14 17:11:28 +08:00
• b41e7cd899 log seq2 benchmark experiments the magician 2026-07-14 15:27:52 +08:00
• 61e169fe18 add cumulative prefix benchmark generator the magician 2026-07-14 14:21:23 +08:00
• c8cc9401a4 baseline working model service and perf tooling the magician 2026-07-14 13:57:40 +08:00
• 1902c81fdd fix issue of loading weight Lu Xinlong 2026-06-30 09:55:13 +08:00
• f89bc60d59 fix multiple issues Lu Xinlong 2026-06-26 17:23:55 +08:00
• 810874ddb8 enable prefix caching Lu Xinlong 2026-06-26 13:27:52 +08:00
• c84151eef9 fix issues Lu Xinlong 2026-06-26 12:55:02 +08:00
• 3d62430fd7 调整配置参数 liwei02 2026-06-25 17:36:43 +08:00
• 72aa7e690a Add README and start commands Lu Xinlong 2026-06-23 17:17:22 +08:00
• b5806731e0 some op overhead optimization Lu Xinlong 2026-06-19 11:19:39 +08:00
• 47a4d9e72a fix no reasoning token issue Lu Xinlong 2026-06-18 12:21:05 +08:00
• 3b8a567e9e fix serving issues when requesting real data Lu Xinlong 2026-06-12 17:57:23 +08:00
• 50e3a05fb0 fix incorrect MoE step to ensure decoding speed Lu Xinlong 2026-06-12 11:44:50 +08:00
• 629f878c28 initial commit for qwen3.6-moe adaptation Lu Xinlong 2026-06-12 10:10:49 +08:00
• 365da18436 Add reasoning parser mechanism + qwen3 parser + bugfixes Lu Xinlong 2026-06-10 18:22:29 +08:00
• 4ab36b51d5 Add qwen3_coder tool calling parser Lu Xinlong 2026-06-10 14:38:54 +08:00
• d972854fb7 fix completion token statistic bug when input context is large Lu Xinlong 2026-06-08 15:04:34 +08:00
• c2de1c83b0 Utilize chunked prefill + K-tiling techniques to ensure 100K context Lu Xinlong 2026-06-05 17:00:41 +08:00
• 2d1ef50992 chunked prefill support and memory opts Lu Xinlong 2026-06-05 16:03:34 +08:00
• 8c047a70ea some modifications to ensure 50K context input Lu Xinlong 2026-06-04 17:56:29 +08:00
• 1c33ef1355 add paged_attn Lu Xinlong 2026-05-29 16:53:39 +08:00
• 3ef8227384 initial version of adding chunked attention, ensuring 20K context Lu Xinlong 2026-05-29 16:49:33 +08:00
• 0e89906481 Qwen3.6-27B iluvatar bi-v100 adaptation Lu Xinlong 2026-05-21 16:37:24 +08:00
• fad74b701b Update to new version of base image Li Jiashu 2025-10-24 15:45:06 +08:00
• ee04aead1e add dataset and more models Zhang Hao 2025-10-17 16:52:12 +08:00
• 8f07ba339a Update README Li Jiashu 2025-08-29 15:40:07 +08:00
• 37e89f390e update Dockerfile images zhousha 2025-08-25 14:19:36 +08:00
• 99fb9f5cb0 First commit Li Jiashu 2025-08-05 19:02:46 +08:00
• 9efe891f99 添加 README.md lumian 2025-08-04 16:57:34 +08:00