feat(QUANT-QWEN38-27B-NVFP4-ARM): the second 27B NVFP4 artifact is a ModelOpt checkpoint, and nothing in this tree read its config

docs/FEATURES.md

@@ -79,7 +79,9 @@ are our reading of their documented behavior, not measurements.
 | GPTQ | ◐ CPU dequant | ✅ | ✅ | ☐ |
 | MXFP4 compressed-tensors | ◐ W4A16 Marlin, mem 2.63x less. gate_up FUSION + decode-graph default-ON; #44 3/3, 32B 6/6. **`VT_MARLIN_DENSE` DEFAULT-ON** (`KERNEL-MARLIN-DENSE-EXEC`): dense marlin 48-CTA, byte-faithful, beats MoE (c8 0.969) | ✅ | ✅ | ☐ |
 | Compressed-tensors `mixed-precision` (`config_groups`, ordered regex `targets`) | ◐ scheme read from the config, never from a dtype probe; an arm with no loader is REFUSED BY NAME. On `unsloth/Qwen3.8-27B-NVFP4` the W4A4 group loads, the FP8 W8A8 group and the `kv_cache_scheme` are refused (#821) | ✅ | ✅ | ☐ |
-| ModelOpt `MIXED_PRECISION`: STATIC per-tensor FP8 + W4A16_NVFP4 | ☐ UNTRIED. `r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121` carries 208 `input_scale`, `F32` scalar `weight_scale` and `"dynamic": false`, so the three `unsloth` blockers are absent (#1574) | ✅ | ✅ | ☐ |
+| ModelOpt `MIXED_PRECISION` (`quantized_layers`, exact module names) | ◐ the declared `quant_algo` is CROSS-CHECKED against the shipped operand names at load; a disagreement, an unimplemented algo or an unknown operand family is REFUSED BY NAME. `r0b0tlab/...-MTP-sm121` loads (#821) | ✅ | ✅ | ☐ |
+| ModelOpt `MIXED_PRECISION`: the cross-check moves no arm | ✅ `nvidia/Qwen3.6-27B-NVFP4` @`0893e160` is NOT refused, and one case says so: same format, wildcard exclusions, an `input_scale` on every NVFP4 module, a `kv_cache_scheme` shipping zero scales (#821) | ✅ | ✅ | ☐ |
+| ModelOpt `MIXED_PRECISION`: STATIC per-tensor FP8 + W4A16_NVFP4 | ◐ LOADS, never RUN. `r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121` carries 208 `input_scale`, `F32` scalar `weight_scale` and `"dynamic": false`, so the three `unsloth` blockers are absent (#1574, #821) | ✅ | ✅ | ☐ |
 | fp8 weights, per-tensor scale | ✅ | ✅ | ✅ | ☐ |
 | Block-wise (fine-grained 128x128) FP8, the `weight_scale_inv` layout | ◐ RUNS on CPU (#1189 M4/M6): 10 projections as 7 GEMMs, `gate_up`+QKV merged. CUDA sm120: N,K %128==0 only; 7 shapes MATCH CPU ref on GB10, no token gate (#1437) ([spec](../.agents/specs/vt-matmul-fp8-block-cuda.md)) | ✅ | ✅ | ☐ |
 | Per-tensor FP8 W8A8 linear is a shared seam any model can bind | ✅ `models/dense_fp8_gemm.h` + `layers::Fp8W8A8LinearMethod` (#940), bound via `layers::MakeLinearMethod`. One definition, CUDA only ([spec](../.agents/specs/vt-fp8-shared-seam.md)) | ✅ `Fp8LinearMethod` | ✅ | ☐ |
@@ -367,6 +369,7 @@ CPU elementwise GEMM (f32/f16/bf16) runs AVX2 and AVX-512 tiers on x86 where the
 | LTX-2.5 resolution | Off-grid sizes refused, naming the offending axis and a size you can actually pass; frames still round | `--width`/`--height` must divide 64 (two-stage) or 32 (one-stage), from the VAE factor times the phase downscale ([#919](https://github.com/mudler/vllm.cpp/issues/919)). `--frames` rounds to `8k + 1`. No size cap |
 | LTX-2.5 arms nothing can request | Declared, not requestable | `int8-convrot` (ComfyUI-only), single-node multi-GPU, `BetaScheduler` (upstream selects no scheduler either). No flag or extra asks for these. `multishot` was RETIRED (absent upstream) and `kLoraFusion` too (now served) |
 | Qwen3.8-27B quantized arms (Q4_K_M GGUF, its `clip` mmproj, the `unsloth` "NVFP4" = `mixed-precision`) | All three files ACCOUNTED against committed manifests (866 / 334 / 1968 names), none RUN: no token gate, Q4_K_M PENDING on #857 | BF16 gated (#915). `--mmproj` (ABI v22) loads a `clip` GGUF; a load refuses a file with a tensor no loader reads. "NVFP4" @`7d6f8d4d`: W4A4 loads, FP8 refused ([spec](../.agents/specs/qwen38-27b-quant-arms.md)) |
+| Qwen3.8-27B, the SECOND NVFP4 artifact (`r0b0tlab/...-MTP-sm121`, a ModelOpt checkpoint) | 2001 names ACCOUNTED per scheme against four committed manifests; LOADS, never RUN: no token gate (#1632) | @`36f717a2`: 208 per-tensor STATIC FP8 + 193 W4A16_NVFP4 modules, both halves load. Not the `unsloth` format ([spec](../.agents/specs/qwen38-27b-quant-arms.md)) |
 | Multi-GPU execution | Hardware-blocked | TP proven equal to tp=1 on CPU; no 2-GPU box to run it |
 | LoRA end to end | CPU brick landed | Unwired standalone; not usable through the server |
 | Multimodal over HTTP | Image request path wired; forward + codec pending | `ROAD-V1-MM` W1-W3 landed. Open: no mm-forward on `Request.mm_features`; no image codec. Video/audio/multi-image now **refuse** with HTTP 400 rather than drop ([#686](https://github.com/mudler/vllm.cpp/issues/686)) |

docs/USAGE.md

@@ -246,6 +246,10 @@ lists other published arms when they have not been used as a gated checkpoint.
 | Qwen3.8-27B GGUF projector | `mmproj-BF16.gguf` | 931,146,432 bytes | `unsloth/Qwen3.8-27B-GGUF` @ `fe1e2a23d973adb629709749dc4f6756df66ef10` | `83ee4f4f205fa514161778c41df1ea14144faa0f713510893b63c2395f5c2d53` | BF16 `clip` projector loads and validates through `--mmproj` | No request path runs the loaded projector |
 | Qwen3.8-27B mixed FP8 and NVFP4 | `model.safetensors` | 22,568,192,096 bytes | `unsloth/Qwen3.8-27B-NVFP4` @ `7d6f8d4d72f56b92b3cdbf22f156b90e1bab0108` | `c473512c70eace07e2256fe9fd76596ac03e3295bee7d54cfb72676416afcc05` | NVFP4 modules load | FP8 modules and quantized KV cache are refused |
 | Qwen3.8-27B MTP drafter | `model_mtp.safetensors` | 849,400,392 bytes | `unsloth/Qwen3.8-27B-NVFP4` @ `7d6f8d4d72f56b92b3cdbf22f156b90e1bab0108` | n/a (non-quantized) | BF16 MTP artifact is present | MTP execution is owed |
+| Qwen3.8-27B ModelOpt NVFP4 shard 1 of 4 | `model-00001-of-00004.safetensors` | 9,965,644,108 bytes | `r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121` @ `36f717a22990e82c54c1d48ee77c491b87825680` | Locally computed hash is owed; the bytes are not mirrored here, and the four shards are verified semantically instead (header parse plus data end equal to the published size); #821 | 208 per-tensor static FP8 and 193 W4A16_NVFP4 modules load | The declared FP8 KV cache is unread; #1593 |
+| Qwen3.8-27B ModelOpt NVFP4 shard 2 of 4 | `model-00002-of-00004.safetensors` | 9,985,743,924 bytes | `r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121` @ `36f717a22990e82c54c1d48ee77c491b87825680` | Locally computed hash is owed; #821 | Same arms as shard 1 | The declared FP8 KV cache is unread; #1593 |
+| Qwen3.8-27B ModelOpt NVFP4 shard 3 of 4 | `model-00003-of-00004.safetensors` | 1,120,886,516 bytes | `r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121` @ `36f717a22990e82c54c1d48ee77c491b87825680` | Locally computed hash is owed; #821 | Same arms as shard 1 | The declared FP8 KV cache is unread; #1593 |
+| Qwen3.8-27B ModelOpt MTP drafter | `model-00004-of-00004.safetensors` | 849,400,592 bytes | `r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121` @ `36f717a22990e82c54c1d48ee77c491b87825680` | Locally computed hash is owed; #821 | Fifteen BF16 MTP tensors are present and unquantized | MTP execution is owed |
 | Qwen3.8-2.4T-A95B | `UD-Q1_0` ten-file GGUF split | about 370 GiB | `unsloth/Qwen3.8-2.4T-A95B-GGUF` @ `567d3e6ac26c5474b18311e619c04350fb9a5556` | `b7770552b2ac24e7334c917bc92e90e218e87cfe29484db65e62e8ef2a60334d` (shard 1); `2765517f833c736338d3ab34354e1c10eb8d79e62325f998285b435e5cf03dcd` (shard 2) | CPU expert streaming from disk | CUDA refuses a checkpoint that exceeds device capacity |
 <!-- checkpoint-registry:end -->