Repository navigation

[Feature] Triton INT4 / INT2 per-token-head KV cache quantization - #40633

Closed

JartX wants to merge 7 commits into

vllm-project:mainfrom

JartX:feature/int4-int2-per-token-head-v2

Closed

[Feature] Triton INT4 / INT2 per-token-head KV cache quantization#40633
JartX wants to merge 7 commits into
vllm-project:mainfrom
JartX:feature/int4-int2-per-token-head-v2

Commits on Apr 24, 2026
- feat: INT4 / INT2 per-token-head KV cache quantization
  JartX
  committed
  d485ea1
- refactor(triton_quant_kv): split packed_per_token_head.py
  JartX
  committed
  b3a14c6
Commits on Apr 25, 2026
- Merge branch 'main' into feature/int4-int2-per-token-head-v2
  JartX
  authored
  d67bcf4
Commits on Apr 28, 2026
- Merge branch 'main' into feature/int4-int2-per-token-head-v2
  JartX
  authored
  8e129ec
Commits on May 4, 2026
- Merge branch 'main' into feature/int4-int2-per-token-head-v2
  JartX
  authored
  48708c8
Commits on May 12, 2026
- fix: INT8 per-token-head KV cache quantizer rounds to nearest
  JartX
  committed
  47de506
Commits on May 14, 2026
- perf(rdna3): dtype-aware RHT sign caching in _hadamard.py
  JartX
  committed
  4d33ce1