1Cat-vLLM

minimax_h3

[Bugfix] Fix SM70 wheel packaging and queued video cleanup

2026-09-29 19:37:37 +08:00

native_media

[Doc] Record real V100 media and progress parity acceptance

2026-09-09 20:47:23 +08:00

arch_overview.md

…

attention_backends.md

[Feature][SM75] Build and load a Turing FlashAttention-2 library next to the Volta one

2026-10-03 05:36:35 +08:00

cuda_graphs_multimodal.md

…

cuda_graphs.md

…

custom_op.md

…

dbo.md

…

debug_vllm_compile.md

…

early_graph_model_inputs.md

[Doc] Record same-engine early graph input comparison

2026-10-05 12:56:44 +08:00

environment_metadata.md

[CI/Build] Register native environment controls and check their readers

2026-10-03 02:25:29 +08:00

flashnext_gguf_graph_parity.md

[Doc] Audit newly merged dense GGUF capabilities

2026-10-05 15:38:15 +08:00

fused_moe_modular_kernel.md

…

fusions.md

…

gguf_dflash2_iq3s.md

[Doc] Compare matched GGUF and NVFP4 acceptance estimators

2026-10-04 19:16:49 +08:00

gguf_fp16_projection.md

[Doc] Record primary GGUF floating row verification

2026-10-05 15:21:24 +08:00

gguf_head_layout_preparation.md

[Doc] Record final Flash-Next graph and acceptance comparison

2026-10-05 15:33:19 +08:00

gguf_iq3_gated_pair_dispatch.md

[Doc] Record GGUF pair model compilation and natural output

2026-10-05 14:29:38 +08:00

gguf_iq3_xxs_pair_reader.md

[Doc] Record installed shared-activation traffic counters

2026-10-05 15:39:26 +08:00

gguf_iq3_xxs_pair_records.md

[Core] Prepare source-sized IQ3_XXS pair records

2026-10-05 14:45:45 +08:00

gguf_iq3xxs_joint_experts.md

[Doc] Round the retained expert bank size consistently

2026-10-05 11:44:32 +08:00

gguf_iq4_native_reader.md

[Kernel] Preserve IQ4_XS source bytes and nested scales (#950 )

2026-10-05 11:11:01 +08:00

gguf_lm_head_sm70.md

[Kernel] Route measured GGUF LM heads through canonical SM70 projections

2026-10-04 19:08:28 +08:00

gguf_mixed_projection_output.md

[Doc] Qualify mixed output with cache-exceeding weights

2026-10-05 12:16:31 +08:00

gguf_native_iq2_s_pair.md

[Kernel] Add source-sized IQ2_S gated-pair reader

2026-10-05 17:52:28 +08:00

gguf_native_iq3xxs_iq4_pair.md

[Kernel] Prepare measured IQ3_XXS and IQ4_XS model admission

2026-10-05 17:42:34 +08:00

gguf_native_mixed_pair.md

[Doc] Record primary V100 mixed pair measurements

2026-10-05 15:20:08 +08:00

gguf_native_q4_k_pair.md

[Doc] Record primary admitted-pair regression and byte coverage

2026-10-05 17:27:02 +08:00

gguf_native_sm70.md

[Doc] Define GGUF fallback scope and current validation

2026-10-03 11:13:57 +08:00

gguf_ordered_ple_rows.md

[Doc] Validate ordered PLE row decoding and graph cost

2026-10-05 15:39:53 +08:00

gguf_packed_ple.md

[Doc] Record installed cascade and real packed row checks

2026-10-05 00:59:52 +08:00

gguf_q4_k_pair_records.md

[Kernel] Preserve original Q4_K bytes in pair records

2026-10-05 15:50:51 +08:00

gguf_qwen35_moe_adapter.md

[Model] Add native Qwen3.5 MoE GGUF adapter

2026-10-05 01:03:34 +08:00

gguf_qwen38_iq3s_layer_types.csv

[Doc] Normalize GGUF inventory line endings

2026-10-05 11:45:49 +08:00

gguf_qwen38_iq3s_source_inventory.md

[Doc] Include all GGUF layer type shares

2026-10-05 11:45:27 +08:00

gguf_turbomind_bitplanes.md

[Doc] Record installed GGUF bit-plane wheel verification

2026-10-03 16:52:13 +08:00

gguf_turbomind_codebook_words.md

[Doc] Record installed lattice word-copy validation

2026-10-04 04:01:31 +08:00

gguf_turbomind_iq3_packet.md

[Bench] Identify loaded GGUF core and retain rejected IQ3 evidence

2026-10-04 03:27:23 +08:00

gguf_turbomind_lattice_banks.md

[Doc] Record installed IQ3 vector codebook results

2026-10-04 04:51:00 +08:00

gguf_turbomind_lattice_decode.md

[Doc] Record matched lattice capture measurements

2026-10-04 01:38:55 +08:00

gguf_turbomind_lattice_grouped.md

[Doc] Record installed GGUF grouped dispatch measurements

2026-10-03 23:48:52 +08:00

gguf_turbomind_lattice.md

[Doc] Record installed GGUF lattice validation

2026-10-03 21:56:00 +08:00

gguf_turbomind_lut4.md

[Doc] Record LUT4 artifact validation on current main

2026-10-03 18:59:41 +08:00

gguf_turbomind_model_moe.md

[Doc] Record installed canonical expert integration checks

2026-10-05 04:28:33 +08:00

gguf_turbomind_model.md

[Doc] Record corrected GGUF model quality and matched C4 repeat

2026-10-04 11:05:25 +08:00

gguf_turbomind_prefill.md

[Doc] Record prefill and LUT4 integration checks

2026-10-03 19:20:20 +08:00

gguf_turbomind_u2.md

[Kernel] Complete grouped GGUF operator comparisons and artifact checks

2026-10-03 14:21:00 +08:00

gguf_turbomind.md

[Doc] Record steady affine timings and installed wheel validation

2026-10-03 13:23:32 +08:00

gguf_v100_control.md

[Model] Integrate standalone Qwen4Exp GGUF mapping on main

2026-10-05 00:47:24 +08:00

grouped_cpu_offload.md

[Bugfix] Retain recent recurrent offload checkpoints by default

2026-10-03 05:40:58 +08:00

h3_encoder_prepared_cache.md

[Core] Reuse exact DiT checkpoint tensors before layout preparation

2026-10-04 02:08:28 +08:00

huggingface_integration.md

…

hybrid_kv_cache_manager.md

…

io_processor_plugins.md

…

logits_processors.md

…

lora_resolver_plugins.md

…

metrics.md

…

mixed_prefill_latency.md

[Doc] Record the final mixed-load replay-window measurements

2026-10-04 02:01:26 +08:00

mm_processing.md

…

model_runner_v2.md

…

moe_kernel_features.md

…

multiprocessing.md

…

nixl_kv_cache_lease.md

…

omni_workflow_coverage.md

[Model] Add native FastH3 Dense fusion and four-step API

2026-09-08 21:20:01 +08:00

optimization_levels.md

[Docs] Update server entrypoint examples (#42077 )

2026-05-09 02:03:52 +00:00

p2p_nccl_connector.md

…

paged_attention.md

…

ple_mapped_row_gather.md

[Kernel] Prefetch and gather disk-mapped PLE rows on CPU (#885 )

2026-10-05 12:18:47 +08:00

ple_metadata_request_indices.md

[Doc] Record metadata transfer hits in the full model

2026-10-05 15:33:20 +08:00

ple_split_mainline_audit.md

[Bugfix][PLE] Preserve padded shards and initialize dummy split tables

2026-09-07 11:14:02 +08:00

plugin_system.md

[Deprecation][1/2] Remove items deprecated in v0.18 (#36470 )

2026-03-09 03:43:23 -07:00

prefix_caching.md

…

qwen4exp_ple_tier_cascade.md

[Bugfix] Admit PLE offload with decode context parallelism

2026-10-04 03:11:55 +08:00

qwen4exp_qsa_dcp2.md

[Doc][QSA] Document the Qwen4Exp QSA DCP2 design and validation

2026-10-03 13:10:45 +08:00

qwen36_27b_awq_mtp_target_verifier_fastpaths.md

…

safetensors_direct_io.md

[Model Loader] Admit direct reads by storage and memory capacity

2026-10-03 08:07:30 +08:00

sm70_awq_exact_m5_batched_gemv.md

…

sm70_awq_fp8_mtp_experts.md

[Bugfix][Qwen4Exp] Allow checkpoint FP8 MTP experts under pipeline parallelism

2026-09-14 22:28:10 +02:00

sm70_awq_long_prefill_exact_dense.md

…

sm70_awq_qpn_m1.md

[Perf][SM70] Enable admitted AWQ QPN M1 with safe fallback

2026-09-06 15:24:47 +08:00

sm70_awq_small_n_hmma_operator.md

…

sm70_concurrency_integration_audit.md

[Core][SM70] Compose concurrency paths with current E4M3 precision main

2026-09-07 18:46:37 +08:00

sm70_deepseek_v4_dspark_acceptance.md

…

sm70_deepseek_v4_dspark_n7_latest_main.md

…

sm70_deepseek_v4_dspark_optimization.md

…

sm70_deepseek_v4_dspark_verifier_20ms_long_context.md

…

sm70_deepseek_v4_flash_bringup.md

…

sm70_deepseek_v4_fp16_gemv.md

…

sm70_deepseek_v4_mhc_decode.md

…

sm70_deepseek_v4_mxfp4_active_expert.md

…

sm70_deepseek_v4_mxfp4_grouped_decode.md

…

sm70_deepseek_v4_performance_integration.md

…

sm70_deepseek_v4_sparse_mla_qk_dsplit.md

…

sm70_deepseek_v4_sparse_mla_splitk.md

…

sm70_deepseek_v4_tp8_graph_allreduce.md

…

sm70_dflash2_acceptance_20260909.md

[Doc] Clarify the included prefill dependency

2026-09-09 20:51:44 +08:00

sm70_dflash2_acceptance_rd.md

…

sm70_dflash2_attention_resources_20260910.md

[Bench] Screen exact QK reuse toward the 7-ms attention target

2026-09-10 14:12:05 +08:00

sm70_dflash2_batch_attention_gdn.md

[Core] Complete automatic DFlash verifier prerequisites

2026-09-25 21:43:23 +08:00

sm70_dflash2_batch_latency_20260926.md

[Doc] Record DFlash2 merge disposition and remaining gates

2026-09-27 02:11:13 +08:00

sm70_dflash2_context_cost_20260909.md

[Perf] Record restored GDN prefill and guard route selection

2026-09-09 23:01:43 +08:00

sm70_dflash2_fastpath_numerics.md

[Bugfix] Preserve proposal probabilities in DFlash2 lookup fusion

2026-09-06 14:36:52 +08:00

sm70_dflash2_fp8_verify_20ms.md

[Doc] Retain the frozen DFlash2 verification baseline

2026-10-02 23:22:27 +08:00

sm70_dflash2_fp32_defaults.md

[Bugfix] Unify DFlash2 E4M3 attention on FP32 state

2026-09-08 09:02:17 +08:00

sm70_dflash2_long_batch_20260927.md

[Doc] Record DFlash2 merge disposition and remaining gates

2026-09-27 02:11:13 +08:00

sm70_dflash2_long_decode_followup_20260927.md

[Doc] Record approval to integrate the reviewed decode changes

2026-09-27 09:23:42 +08:00

sm70_dflash2_long_verify_curve_20260909.md

[Bench] Validate visible q8 rounds and isolate tail attention costs

2026-09-10 09:44:09 +08:00

sm70_dflash2_long_verify_decay.md

…

sm70_dflash2_lookup_augmented.md

[Bugfix] Preserve proposal probabilities in DFlash2 lookup fusion

2026-09-06 14:36:52 +08:00

sm70_dflash2_m48_scaling_20260927.md

[Doc] Record main integration and merge validation for M48

2026-09-27 23:58:50 +08:00

sm70_dflash2_ngram_hybrid.md

…

sm70_dflash2_nvfp4_17ms.md

…

sm70_dflash2_nvfp4_prefill_promotion.md

…

sm70_dflash2_policy.md

[Core] Retire failed DFlash2 candidate-order override

2026-10-03 08:09:52 +08:00

sm70_dflash2_prefill_closure.md

…

sm70_dflash2_quality_audit.md

…

sm70_dflash2_retired_candidate_order.md

[Core] Retire failed DFlash2 candidate-order override

2026-10-03 08:09:52 +08:00

sm70_dflash2_shared_nvfp4.md

[Doc] Record SM70 memory savings and matched graph validation

2026-09-22 11:03:49 +08:00

sm70_dflash2_tail_graphs_20260911.md

[Perf] Ship the qualified SM70 long-context layout and drop its context fences

2026-09-12 20:53:06 +08:00

sm70_dflash2_target_graph_20ms.md

…

sm70_dflash2_tp2_verifier.md

[Doc] Pin the accepted TP2 head and integration switches

2026-09-10 00:26:33 +08:00

sm70_dflash2_verifier_route_audit.md

[Core] Integrate DFlash2 quality repairs with current main

2026-09-06 19:11:53 +08:00

sm70_dflash_ddtree_27b_awq_plan.md

…

sm70_dflash_smallm.md

[Kernel] Add packed small-M DFlash2 projections on SM70

2026-10-04 13:11:35 +08:00

sm70_e4m3_grouped_fp32.md

[Bugfix] Unify DFlash2 E4M3 attention on FP32 state

2026-09-08 09:02:17 +08:00

sm70_environment_surface.md

[Core] Unify SM70 verifier trace and selector shadow controls

2026-10-03 08:21:12 +08:00

sm70_exact_auxiliary_fusion.md

[Kernel] Isolate unused MoE and auxiliary policies from compile caches

2026-10-03 19:42:44 +08:00

sm70_fa2_d256_prefill_pipeline.md

…

sm70_flash_next_batch_defaults.md

[Bugfix] Align FP32 router preparation with its runtime precision guard

2026-10-05 01:34:19 +08:00

sm70_flash_next_mtp4_256k_eval_20260927.csv

[Doc] Normalize measurement CSV line endings

2026-10-02 23:22:28 +08:00

sm70_flash_next_mtp4_256k_eval_20260927.md

[Doc] Retain the frozen Flash-Next accuracy and capacity measurements

2026-10-02 23:22:28 +08:00

sm70_flash_next_mtp4_batch_gdn.md

[Doc] Record MTP4 merge audit and integration gates

2026-09-27 18:57:53 +08:00

sm70_flash_next_mtp4_default_profile.md

[Bugfix][PLE] Defer hybrid cache geometry until cache setup

2026-10-03 01:23:30 +08:00

sm70_flash_next_mtp4_trace_reconciliation.md

[Doc] Keep timing reports focused on reproducible measurement contracts

2026-10-02 23:22:27 +08:00

sm70_flash_next_prefix_prefill.md

[Bugfix][Scheduler] Limit full-chunk Mamba fallback to contention

2026-10-03 01:04:45 +08:00

sm70_flash_v37_prefill.md

[Core] Enable qualified SM70 75T prefill by default

2026-09-14 17:20:01 +08:00

sm70_flash_v100_fp8_kv_long_context.md

[Core][SM70] Capture concurrent FP8 KV graphs through B32

2026-09-15 01:30:32 +08:00

sm70_flash_v100_prefill_operator_optimization.md

…

sm70_flashinfer_native_backend.md

…

sm70_flashnext_gguf_mtp4_operator_routes.md

[Benchmark] Record reproducible Flash-Next GGUF MTP4 composition

2026-10-05 06:13:17 +08:00

sm70_flashnext_gguf_mtp4.md

[Doc] Record the repeated Flash-Next C4 ring comparison

2026-10-04 17:25:47 +08:00

sm70_fp8_kernel_integration.md

[Core] Move SM70 serialized FP8 dispatch into the linear kernel lifecycle

2026-10-03 03:36:20 +08:00

sm70_fp8_long_prefill_exact_dense.md

…

sm70_fp8_policy_cleanup.md

[Core] Remove redundant SM70 FP8 startup environment writes

2026-10-03 01:32:11 +08:00

sm70_fp16_gemv_silu_schedule.md

[Kernel] Improve small-grid FP16 GEMV SiLU scheduling on SM70 (#812 )

2026-10-03 11:06:40 +08:00

sm70_fp16_grouped_verify.md

[Doc] Record packaged FP16 grouped backend validation

2026-10-04 20:02:28 +08:00

sm70_gdn_ba_verify.md

[Kernel] Fuse SM70 verifier b/a and channel FP8 projections

2026-10-04 13:33:18 +08:00

sm70_gdn_preprocess.md

[Doc] Record five-row GDN preprocessing correctness and timing

2026-10-05 02:40:34 +08:00

sm70_gdn_projection_copy_audit.md

[Kernel][SM70] Fuse copy-only batched GDN projection splits

2026-09-07 18:14:47 +08:00

sm70_gdn_projection_tails.md

[Doc] Record default GDN tail-copy route and timing

2026-10-05 05:17:04 +08:00

sm70_gdn_verify_tile.md

[Kernel] Tile single-request SM70 GDN verification columns

2026-10-04 12:51:02 +08:00

sm70_gemma_long_prefill_fusion.md

…

sm70_gguf_canonical_head.md

[Doc] Record packed LM-head method measurements

2026-10-04 21:32:01 +08:00

sm70_gguf_expert_down_dispatch.md

[Doc] Record installed GGUF down dispatch validation

2026-10-05 04:33:07 +08:00

sm70_gguf_fp16_dense_routes.md

[Doc] Record extended GDN row dispatch checks

2026-10-04 21:31:28 +08:00

sm70_gguf_joint_expert_dispatch.md

[Doc] Record joint expert model dispatch qualification

2026-10-05 06:11:25 +08:00

sm70_gguf_joint_original_experts.md

[Doc] Qualify standalone joint experts with cold-bank timings

2026-10-05 06:10:46 +08:00

sm70_gguf_small_grouped_down.md

[Kernel] Admit measured small grouped down vector batches

2026-10-05 01:12:07 +08:00

sm70_glm53_flash_nvfp4.md

[Bugfix] Unify DFlash2 E4M3 attention on FP32 state

2026-09-08 09:02:17 +08:00

sm70_hc_batch_topology.md

[Doc] Record M20 HC calibration and bounded admission

2026-10-05 02:15:38 +08:00

sm70_hc_weight_prefetch_screen.md

[Benchmark] Record reproducible Flash-Next GGUF MTP4 composition

2026-10-05 06:13:17 +08:00

sm70_hmma_pipeline_search.md

…

sm70_linear_kernel_integration.md

[Kernel] Select Turing QPN linears through kernel capabilities

2026-10-03 06:39:47 +08:00

sm70_loaded_route_reporting.md

[Core] Report loaded SM70 kernel decisions for every model

2026-10-03 06:22:18 +08:00

sm70_mamba_state_grid_decoupling.md

[Release][SM70] Ship portable V100 launcher and wheel gate

2026-09-29 23:08:03 +08:00

sm70_memory_arena.md

[Memory] Reduce SM70 graph scratch and DFlash active peak

2026-09-23 09:58:34 +08:00

sm70_memory_defaults.md

[Doc] Record same-build shared-weight probability parity

2026-09-22 11:08:34 +08:00

sm70_memory_workspace_tuning.md

[Memory] Reduce SM70 graph scratch and DFlash active peak

2026-09-23 09:58:34 +08:00

sm70_mixed_batch_decode_rows.md

[Perf][SM70] Keep resident decode rows on fast routes inside mixed prefill batches

2026-10-03 03:18:46 -07:00

sm70_mrv2_sampling_contract_repairs.md

[Bugfix][MRv2] Repair six sampling and logprob runtime defects (#533 )

2026-09-07 01:07:54 +08:00

sm70_mtp_gdn_output.md

[Bench] Record direct MTP recurrent output copy savings

2026-10-05 03:40:28 +08:00

sm70_mtp_strided_qkv.md

[Bench] Record strided MTP QKV timing and exact state validation

2026-10-05 03:25:52 +08:00

sm70_nvfp4_turbomind_operator_optimization.md

…

sm70_ple_ngram_batch.md

[Doc] Record exact PLE batch indices and graph operator savings

2026-10-05 00:10:00 +08:00

sm70_prefill_score_overflow.md

[Bugfix] Recover overflowing SM70 compact score tiles

2026-10-03 21:12:09 +08:00

sm70_prefill_tail_intermediate.md

[Doc] Record long-context tail peak recovery

2026-10-04 10:05:19 +08:00

sm70_prefix_anchored_swa.md

…

sm70_prefix_peak_repair.md

[Bugfix] Repair missed SM70 prefill prefix maxima

2026-10-03 20:38:28 +08:00

sm70_qsa_page4_allocation_invariance.md

[Bugfix][SM70] Stabilize QSA page4 order under KV relocation

2026-09-06 14:26:41 +08:00

sm70_quantized_batch_reuse.md

[Kernel] Enable shared SM70 C2 C4 C8 batch acceleration by default (#691 )

2026-09-26 18:43:15 +08:00

sm70_quasar_dflash2_15ms.md

[Perf] Validate bound attention and screen chunked QPN2

2026-09-09 01:30:59 +08:00

sm70_quasar_dflash2_operator_audit.md

[Kernel] Add E4M3 KV and FP32 logits for the DFlash2 quality path

2026-09-06 10:20:17 +08:00

sm70_quasar_dflash2_quality_speed_recovery.md

[Doc] Record production confirmation of DFlash2 latency recovery

2026-09-06 12:01:50 +08:00

sm70_quasar_dflash2_resource_audit_20260909.md

[Perf] Separate common DFlash2 schedules from NVFP4 projection loading

2026-09-09 18:04:08 +08:00

sm70_quasar_e4m3_fp32_logits.md

[Kernel] Add E4M3 KV and FP32 logits for the DFlash2 quality path

2026-09-06 10:20:17 +08:00

sm70_quasar_nvfp4_dflash2_acceptance.md

…

sm70_quasar_tp2_tp4_quality.md

[Audit] Trace QUASAR TP partition drift through target sampling

2026-09-06 14:01:26 +08:00

sm70_qwen36_27b_awq_mtp4_optimization.md

…

sm70_qwen38_130_acceptance.md

…

sm70_qwen38_awq_active_grouped_decode.md

docs: correct engine-reported AWQ KV capacity

2026-09-05 10:58:23 -07:00

sm70_qwen38_concurrency40.md

[Doc] Record repaired Qwen3.8 quality and concurrent baselines

2026-09-28 01:15:53 +08:00

sm70_qwen38_default_fastpath.md

[Bugfix][PLE] Budget grouped caches and preserve split-K controls

2026-10-02 22:33:53 +08:00

sm70_qwen38_distribution_acceptance.md

[Kernel] Prefetch and gather disk-mapped PLE rows on CPU (#885 )

2026-10-05 12:18:47 +08:00

sm70_qwen38_exact_decode_followup.md

[CI] Clear NVFP4 integration checks without changing runtime arithmetic

2026-09-06 16:57:36 +08:00

sm70_qwen38_flash_next_nvfp4.md

…

sm70_qwen38_fp8_prefill_5500.md

…

sm70_qwen38_fp8_prefill_decay.md

…

sm70_qwen38_nomtp_concurrency.md

[Perf][WIP] Gate HC sharding on rotating-weight performance

2026-09-05 15:08:29 +08:00

sm70_qwen38_nvfp4_decode.md

[Doc] Record registered HC fusion quality and full-chain gain

2026-09-05 15:32:03 +08:00

sm70_qwen38_nvfp4_quality_repair.md

[CI] Clear NVFP4 integration checks without changing runtime arithmetic

2026-09-06 16:57:36 +08:00

sm70_qwen38_ple_disk_offload.md

[Perf] Integrate Flash-Next MTP4 acceleration defaults

2026-09-26 18:36:47 +08:00

sm70_qwen38_projection_w13_followup.md

[Bugfix][SM70] Preserve projection roles in M1 GEMV dispatch

2026-09-06 00:40:16 +08:00

sm70_qwen38_qpn8_decode.md

…

sm70_qwen38_reproducible_baseline.md

[Doc] Document Qwen3.8 default-route selection and validation

2026-09-09 12:00:14 +08:00

sm70_qwen38_step0_kernel_budget.md

[Perf] Add decode boundary and PLE budget measurements

2026-10-04 11:30:32 +08:00

sm70_release151_compile_cache.md

[Bugfix][SM70] Reuse compiled graph caches for E4M3 DFlash2

2026-10-02 19:11:58 +08:00

sm70_sampling_input_guards.md

[Doc] Describe model-aware sampling input validation

2026-10-03 06:53:40 +08:00

sm70_sm120_long_context_attention_parity.md

…

sm70_small_moe_routing.md

[Doc] Record exact expert restoration and graph timing

2026-10-04 21:52:40 +08:00

sm70_sparse_bmm_capabilities.md

[Perf] Preserve faster paged lanes and avoid padded indexer graph work

2026-10-03 10:39:32 +08:00

sm70_tile_runtime_exploration.md

…

sm70_tp4_mtp4_multigpu_scaling.md

…

sm70_tp4_nomtp_long_context_decode.md

…

sm70_tp4_push_gemma_norm.md

[Kernel] Fuse SM70 TP4 push allreduce and Gemma norm

2026-10-04 12:04:13 +08:00

sm70_tp_quality_audit.md

[Test] Record matched quality audits and cutoff coverage

2026-09-22 00:44:17 +08:00

sm70_tp_shape_coverage.md

[Doc] Record paired quality audit and QPN2 precision repair

2026-09-21 22:30:09 +08:00

sm70_v100_migration_control.md

[Core] Stage PLE results in mapped host buffers

2026-10-03 14:04:54 +08:00

speculative_draft_load_config.md

[Doc] Record installed draft loader regression checks

2026-10-05 00:44:42 +08:00

torch_compile_multimodal.md

…

torch_compile.md

…

vllm_ir.md

…