AlpinDale 7ca63930c8 support deepseek_v3 model пре 1 месец
..
fused_moe 7ca63930c8 support deepseek_v3 model пре 1 месец
mamba f1d0b77c92 [0.6.0] Release Candidate (#481) пре 5 месеци
ops 8a71788372 Add OLMoE (#772) пре 3 месеци
__init__.py 07aa2a492f upstream: add option to specify tokenizer пре 1 година
activation.py f1d0b77c92 [0.6.0] Release Candidate (#481) пре 5 месеци
layernorm.py bfc3da41ae feat: add torch.compile for GemmaRMSNorm (#898) пре 1 месец
linear.py cec4da1dab quants: support w8a8 fp8 block-wise quantization from DS3 пре 1 месец
logits_processor.py 0e558e9b2f fix: loading chameleon model with TP>1 (#695) пре 4 месеци
pooler.py f1d0b77c92 [0.6.0] Release Candidate (#481) пре 5 месеци
rejection_sampler.py e3a53712f2 fix: mlpspeculator with padded vocab (#669) пре 4 месеци
resampler.py 548e864404 models: add support for QwenVL (#995) пре 1 месец
rotary_embedding.py d51720114b chore: use RoPE cache for MRoPE method (#1028) пре 1 месец
sampler.py 2261a0e8dd cpu: fix issue with sampling kernels (#1016) пре 1 месец
spec_decode_base_sampler.py 09b82f9963 feat: Add support for GPU device selection in SpecDecodeBaseSampler (#629) пре 5 месеци
typical_acceptance_sampler.py f1d0b77c92 [0.6.0] Release Candidate (#481) пре 5 месеци
vocab_parallel_embedding.py 83af2524f3 quants: add GPTQ and FBGEMM to AphroditeParameters (#987) пре 1 месец