HAI
|
e11ab79e68
|
[Performance, hardware] MoE tuning update to AMD MI300x GPUs (#1619)
|
2024-10-10 22:48:15 -07:00 |
|
Byron Hsu
|
01fdb2f377
|
Fix test_vision_openai_server on CI (#1620)
|
2024-10-10 16:34:13 -07:00 |
|
Amos You
|
c996e8ccd4
|
[Minor] Fix logging typo (#1615)
|
2024-10-08 21:11:19 -07:00 |
|
Lianmin Zheng
|
7b69d91b4f
|
Release v0.3.3 (#1605)
|
2024-10-08 12:58:41 -07:00 |
|
Byron Hsu
|
e8613df071
|
[Engine] Fix generate hanging issue after the first call (#1606)
|
2024-10-08 04:26:56 +00:00 |
|
Ying Sheng
|
c5325aba75
|
[Profile] Add pytorch profiler (#1604)
|
2024-10-07 14:37:16 -07:00 |
|
Lianmin Zheng
|
ebbc42d989
|
Optimize broadcast & Reorg code (#1598)
|
2024-10-07 13:19:23 -07:00 |
|
Jani Monoses
|
3ff641132e
|
Remove references to squeezellm (#1603)
|
2024-10-07 11:30:41 -07:00 |
|
Lianmin Zheng
|
2b302b9393
|
Fix the port_args in bench_latency (#1597)
|
2024-10-07 00:44:38 -07:00 |
|
Ke Bao
|
68f8b60d22
|
Fix chunked prefill condition (#1594)
|
2024-10-07 06:34:14 +00:00 |
|
Lianmin Zheng
|
6a5b352aaf
|
Use is_flashinfer_available to replace is_hip for flashinfer check (#1596)
Co-authored-by: Zhang Liangang <liangang.zhang@intel.com>
|
2024-10-06 22:54:05 -07:00 |
|
Byron Hsu
|
565b05f02f
|
Use atexit hook to implicitly shutdown Runtime (#1595)
|
2024-10-07 05:18:45 +00:00 |
|
Lianmin Zheng
|
b6aad70ab1
|
[Fix] Fix the case where prompt_len = 0 (#1593)
|
2024-10-06 20:30:02 -07:00 |
|
Byron Hsu
|
551a3a9d38
|
Provide an offline engine API (#1567)
|
2024-10-06 20:27:03 -07:00 |
|
Lianmin Zheng
|
91877a9f9c
|
Fix modality for image inputs (#1592)
|
2024-10-06 15:43:32 -07:00 |
|
Ying Sheng
|
c98e84c21e
|
[Minor, Performance] Use torch.argmax for greedy sampling (#1589)
|
2024-10-06 13:15:05 -07:00 |
|
Ying Sheng
|
9c064bf78a
|
[LoRA, Performance] Speedup multi-LoRA serving - Step 1 (#1587)
|
2024-10-06 10:33:44 -07:00 |
|
Lianmin Zheng
|
58d1082e39
|
Clean up event loop (#1586)
|
2024-10-06 03:24:04 -07:00 |
|
HAI
|
4d086719e5
|
[Bug] Fix decode stats error on output_len 1 (#1585)
|
2024-10-06 08:09:09 +00:00 |
|
Lianmin Zheng
|
9244f27f0a
|
[Minor] Improve the style and fix flaky tests (#1584)
|
2024-10-06 00:10:48 -07:00 |
|
Byron Hsu
|
2422de5193
|
Support min_tokens in sgl.gen (#1573)
|
2024-10-05 21:51:12 -07:00 |
|
Byron Hsu
|
521f862d90
|
Fix runtime.generate when sampling param is not passed (#1582)
|
2024-10-05 17:59:05 -07:00 |
|
Byron Hsu
|
34c32d2820
|
Fix styling (#1583)
|
2024-10-05 17:52:14 -07:00 |
|
Byron Hsu
|
dde8bb16fe
|
default sampling param should be deepcopied (#1581)
|
2024-10-05 17:27:43 -07:00 |
|
Byron Hsu
|
8ac3ccc060
|
Backend method not found when SRT Runtime is used (#1576)
|
2024-10-05 11:47:35 -07:00 |
|
Jerry Zhang
|
9b0926ceeb
|
Add llama implementation with no tensor parallel linears (#1561)
|
2024-10-05 11:22:27 -07:00 |
|
Byron Hsu
|
6bfdb4031d
|
[Easy] use .text() instead of .text (#1577)
|
2024-10-05 11:07:41 -07:00 |
|
Liangsheng Yin
|
5d0ba4038f
|
Refine the add request reasons to avoid corner cases. (#1574)
|
2024-10-04 18:00:18 -07:00 |
|
Ying Sheng
|
04b262cd91
|
[Fix] Fix major performance bug in certain cases (#1563)
Co-authored-by: hnyls2002 <hnyls2002@gmail.com>
|
2024-10-04 08:51:11 +00:00 |
|
Lianmin Zheng
|
45473d4b2b
|
Make input_ids a torch.Tensor (#1568)
|
2024-10-04 01:09:59 -07:00 |
|
Lianmin Zheng
|
114bbc8651
|
Use ipc instead of tcp in zmq (#1566)
|
2024-10-04 00:45:52 -07:00 |
|
Lianmin Zheng
|
32eb6e96f2
|
Organize sampling batch info better (#1562)
|
2024-10-03 18:29:49 -07:00 |
|
HAI
|
e0b5dbcec1
|
[FP8 KV Cache] Avoid KeyError at loading pre-quantized FP8 model with kv_scale (#1559)
|
2024-10-03 01:52:26 -07:00 |
|
Minsang Song
|
e6852b0dd2
|
[Fix] Fix AttributeError in Qwen2.5 LoRA: 'Qwen2ForCausalLM' object has no attribute 'get_hidden_dim' (#1536)
Co-authored-by: Ying Sheng <sqy1415@gmail.com>
|
2024-10-02 20:41:15 -07:00 |
|
Lianmin Zheng
|
4ae0969c0a
|
Move status check in the memory pool to CPU (#1557)
|
2024-10-02 18:23:35 -07:00 |
|
Lianmin Zheng
|
317631cada
|
[Fix] Move ScheduleBatch out of SamplingInfo (#1556)
|
2024-10-02 17:18:04 -07:00 |
|
Lianmin Zheng
|
b564835364
|
[Fix] do not maintain regex_fsm in SamplingBatchInfo (#1555)
|
2024-10-02 13:19:44 -07:00 |
|
Theresa Barton
|
2c7d0a5b8b
|
[Fix] Fix all the Huggingface paths (#1553)
|
2024-10-02 10:12:07 -07:00 |
|
kk
|
8cdc76f6d4
|
[Performance, Hardware] MoE tuning on AMD MI300x GPUs (#1554)
Co-authored-by: wunhuang <wunhuang@amd.com>
|
2024-10-02 09:52:46 -07:00 |
|
Ying Sheng
|
f202ed9712
|
[Refactor] Simplify io_struct and tokenizer_manager (#1549)
|
2024-10-01 10:25:32 -07:00 |
|
Liangsheng Yin
|
100f5b8bc9
|
Simplify flashinfer dispatch (#1552)
|
2024-10-01 00:28:42 -07:00 |
|
Liangsheng Yin
|
619bb6ddda
|
Dispatch flashinfer wrappers (#1550)
|
2024-09-30 23:12:36 -07:00 |
|
Liangsheng Yin
|
b88ea90d4a
|
Fix bugs of logprobs_nums (#1548)
|
2024-09-30 17:09:54 -07:00 |
|
Liangsheng Yin
|
99ec439da4
|
Organize Attention Backends (#1547)
|
2024-09-30 15:54:18 -07:00 |
|
Ying Sheng
|
0f4fb19bc8
|
[Fix, LoRA] fix LoRA with updates in main (#1545)
|
2024-09-30 10:06:08 -07:00 |
|
Lianmin Zheng
|
63ba2f8d7b
|
Clean up batch data structures: Introducing ModelWorkerBatch (#1544)
|
2024-09-30 06:41:49 -07:00 |
|
Lianmin Zheng
|
36d5acfca5
|
Rename InputMetadata -> ForwardBatch (#1543)
|
2024-09-30 02:41:11 -07:00 |
|
Lianmin Zheng
|
3f0fe08d37
|
Let ModelRunner take InputMetadata as input, instead of ScheduleBatch (#1541)
|
2024-09-29 20:28:45 -07:00 |
|
Liangsheng Yin
|
55b974f96f
|
Process image in parallel (#1539)
|
2024-09-29 18:52:43 -07:00 |
|
Lianmin Zheng
|
f86c1e611f
|
Move scheduler code from tp_worker.py to scheduler.py (#1538)
|
2024-09-29 17:42:45 -07:00 |
|