Open Issues Need Help
View All on GitHub [FP8] Support block-wise FP8 quantization 12 days ago
enhancement help wanted
[FP8] Add FP8 support to ParallelLMHead 13 days ago
enhancement help wanted September2026
[FP8] Support FP8 KV cache for attention 16 days ago
enhancement help wanted
help wanted
enhancement help wanted September2026
[Feature]: Configure individual layers on spyre 20 days ago
enhancement good first issue
[Epic] Support prefill/decode disaggregation 20 days ago
help wanted
[Epic] Develop KVCacheConnector for Spyre 20 days ago
help wanted
enhancement help wanted
enhancement help wanted September2026
remove cpu-roundtrip for `reshape_and_cache` about 1 month ago
help wanted vllm-spyre-next vllm-spyre-attention-backend August2026
[FP8] Add FP8 test infrastructure and reference tests about 1 month ago
enhancement help wanted vllm-spyre-next tests August2026
[FP8] Fix Inductor compatibility for FP8 tensors (torch-spyre #1474) about 1 month ago
enhancement help wanted vllm-spyre-next August2026
[FP8] Add FP8 tensor initialization kernels for spyre device about 1 month ago
enhancement help wanted vllm-spyre-next August2026
[FP8] Implement scaled_mm operator support in torch-spyre about 1 month ago
enhancement help wanted vllm-spyre-next August2026
add indirect access for matmul in online softmax about 1 month ago
help wanted vllm-spyre-next vllm-spyre-attention-backend August2026
remove CPU round-trip for attention result-reshape about 2 months ago
help wanted vllm-spyre-next vllm-spyre-attention-backend
good first issue vllm-spyre-next July2026
Verify the 4D matmul in the attention computation 4 months ago
good first issue vllm-spyre-next June2026
Write design docs for key features 5 months ago
good first issue May2026
Wrap the ParallelLMHead for torch-spyre 5 months ago
good first issue vllm-spyre-next
[doc] add a README.md 5 months ago
documentation good first issue