repo stringlengths 23 79 | downloads int64 0 1.49M | likes int64 0 346 | last_modified timestamp[s]date 2026-08-14 00:00:00 2026-08-24 00:00:00 | format stringclasses 5
values | ignore_entries int64 0 519 | config_declares_mtp bool 2
classes | artifact_sha stringlengths 64 64 ⌀ | quant_config_sha stringlengths 16 16 | tensors int64 23 3.02k ⌀ | mtp_tensors int64 0 15 ⌀ | mtp_scale_keys int64 0 16 ⌀ | mtp_q_modules int64 0 8 ⌀ | mtp_w_modules int64 0 15 ⌀ | la_proj_modules int64 0 240 ⌀ | la_proj_quantized int64 0 240 ⌀ | visual_tensors int64 0 333 ⌀ | visual_quantized int64 0 330 ⌀ | visual_q_modules int64 0 111 ⌀ | visual_w_modules int64 0 167 ⌀ | linear_attn_tensors int64 0 432 ⌀ | linear_attn_scale_keys int64 0 720 ⌀ | linear_attn_q_modules int64 0 240 ⌀ | linear_attn_w_modules int64 0 336 ⌀ | mtp_quantized int64 0 16 ⌀ | linear_attn_quantized int64 0 720 ⌀ | mtp_in_ignore bool 2
classes | scope stringclasses 3
values | verdict stringclasses 8
values | note stringclasses 22
values | duplicate_of stringclasses 8
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
unsloth/Qwen3.8-27B-NVFP4 | 1,488,621 | 346 | 2026-08-17T00:00:00 | mixed-precision | 303 | true | e73aa0e49431c474f8918c866d4ccbe8fdb10c39edcb193b9916a9460045b26a | c0f8f7d387eb34d0 | 1,968 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
RadixArk/Qwen3.8-27B-NVFP4 | 856,173 | 57 | 2026-08-22T00:00:00 | modelopt | 2 | true | 63fa79a94200915be9c3493a293499947827217175fe5a8ea1d14b801b5b738b | 96a96d28dc135516 | 2,194 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
Inferact/Qwen3.8-27B-NVFP4 | 667,779 | 12 | 2026-08-14T00:00:00 | modelopt | 244 | true | 074bb35853339e2b2c27091ca9ee7819b45e0551458cd6ef2492063768e71e1e | 013ec4449e26eaef | 2,111 | 15 | 0 | 0 | 15 | 240 | 48 | 333 | 0 | 0 | 167 | 432 | 144 | 48 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090 | 111,155 | 113 | 2026-08-19T00:00:00 | modelopt | 148 | true | c4fc71330751f54dc06c7d6b4c5cc51bbd3005c4b518b3bb8c0fb1a6efa55349 | c2eb2f759c3ec670 | 2,402 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
sakamakismile/Qwen3.8-27B-MTP-NVFP4 | 83,253 | 42 | 2026-08-16T00:00:00 | nvfp4-pack-quantized | 222 | true | 7384170b97da8e42de87467571142efb8b3cf560c53f7db9eaeb833d72574a11 | 1414efc09a75f72b | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
sakamakismile/Qwen3.8-27B-AEON-ULTIMATE-UNCENSORED-NVFP4 | 30,927 | 17 | 2026-08-16T00:00:00 | nvfp4-pack-quantized | 222 | true | 51f2c3f886a61ac9e363927bbf0dd13c9afd7be71d442bc386e113541a2116b5 | 1414efc09a75f72b | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
neroued/Qwen3.8-27B-nvfp4-NInfer | 27,733 | 26 | 2026-08-19T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
sakamakismile/Huihui-Qwen3.8-27B-abliterated-NVFP4 | 20,354 | 21 | 2026-08-24T00:00:00 | nvfp4-pack-quantized | 222 | true | 94a8afaed3179eab0ced1bcf380ebc469aa958de801f3fdde0cb719abf1cd832 | 1414efc09a75f72b | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
orcarouter/Qwen3.8-27B-Uncensored-NVFP4 | 19,624 | 62 | 2026-08-21T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
joshebbs/qwen3.8-27b-uncensored-nvfp4-modelopt | 14,974 | 10 | 2026-08-15T00:00:00 | modelopt | 146 | true | 556ecd0433b09da5ed44feaf5a0bc2c396900dbd4baf7b7c7afd9996e295b955 | acf8724ee9644b4b | 2,399 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
Blackfrost-AI/Qwen3.8-27B-ABLITERATED-NVFP4 | 9,375 | 12 | 2026-08-16T00:00:00 | modelopt | 101 | true | d325f6fa39d0f911454a86c03716ca4f23a856bae4cff348d987de60b10a9cf4 | c20d8c2fa376a22f | 1,823 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
hwkranger/Qwen3.8-27B-heretic-ara-NVFP4 | 8,531 | 3 | 2026-08-16T00:00:00 | modelopt | 148 | true | bdb7560b2716a68eeedb4408881c07fba8ce1d931013b12c064aabbe6f43bbde | 86106fb0a7f35bd6 | 1,999 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
huginnfork/Qwen3.8-27B-NVFP4A16 | 6,611 | 20 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 519 | true | 12ae8bcd06704024f0011287efb3d09b9630f0060d1825a62449c383a6026b6c | 765688b61d8368e7 | 1,583 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
gittensor-model-hub/Qwen3.8-27B-DSpark-NVFP4 | 5,802 | 13 | 2026-08-19T00:00:00 | modelopt | 20 | false | 02dcc3ab8115a40260484d820ee03740c2065e06593d39057cd032d8c8ca6bac | 4bcdcb2f672c1c5c | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
aday777/Qwen3.8-27B-ARA-abliterated-NVFP4-MTP | 5,173 | 2 | 2026-08-15T00:00:00 | nvfp4-pack-quantized | 222 | true | db37463963e693706a54e167e5e7425f9757eb388bafd97e15f53b760aaca4bf | cc738c2271671595 | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
kelnei/Qwen3.8-27B-NVFP4 | 5,146 | 0 | 2026-08-17T00:00:00 | mixed-precision | 314 | true | 5f3af5bcb7019181b965de2e5663db418b86a714c855cedcc88cc740cbd5a1e5 | 4baea8e6a72c127d | 1,968 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
rdtand/Qwen3.8-27B-PrismaAQUA-5.5bit-vllm | 4,869 | 26 | 2026-08-23T00:00:00 | mixed-precision | 162 | true | ac9f87096fdcdfade238eafe94931a59fe2a34e93666fb4819716df45d37dc41 | cbd11f92c756ba12 | 2,163 | 15 | 0 | 0 | 15 | 240 | 214 | 333 | 0 | 0 | 167 | 432 | 372 | 214 | 336 | 0 | 372 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
mlx-community/Qwen3.8-27B-nvfp4 | 4,729 | 5 | 2026-08-14T00:00:00 | null | 0 | true | e04a8b70c7c3b5396f27591f725abc304ad3d072933ca5d7ed8e2f3765d615d3 | 9c16fdf1af1c2cbf | 1,682 | 0 | 0 | 0 | 0 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 240 | 240 | 336 | 0 | 240 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
RedHatAI/Qwen3.8-27B-NVFP4 | 3,989 | 4 | 2026-08-17T00:00:00 | mixed-precision | 303 | true | e73aa0e49431c474f8918c866d4ccbe8fdb10c39edcb193b9916a9460045b26a | c0f8f7d387eb34d0 | 1,968 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | duplicate-artifact | every weight shard sha256-identical to unsloth/Qwen3.8-27B-NVFP4, same quantization_config | unsloth/Qwen3.8-27B-NVFP4 |
lyf/Qwen3.8-27B-Blackfrost-Abliterated-NVFP4-MTP-VL | 3,780 | 1 | 2026-08-16T00:00:00 | nvfp4-pack-quantized | 5 | true | 491b4fd36fbd39eb628e1ff22433a9e00b28ed6eaf4b47200fbab7dc8472c2ff | 12f59baff0d71a1d | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
cloudnathan5/Qwen3.8-27B-NVFP4a16-AWQ | 2,536 | 0 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 304 | true | 7658ffe14c0cef8824241b8fd11fc47b50b0a8c04c3532520fab5f99eddacd76 | 932e3879fc25a2ab | 1,999 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
mlx-community/Qwen3.8-27B-MTP-nvfp4 | 2,485 | 1 | 2026-08-14T00:00:00 | null | 0 | true | a0b1689dfd7ae18a93cf8877b1a918936b07445a0960bd6a84cbf1b77b51577f | 9c16fdf1af1c2cbf | 23 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | false | unrecognized | unrecognized-layout | none of visual.*/linear_attn.*/mtp.* found; this checkpoint does not use transformers Qwen3_5 tensor names (MLX and similar re-keyed formats) | null |
lyf/Qwen3.8-27B-Heretic-ARA-NVFP4-MTP-VL | 2,307 | 1 | 2026-08-16T00:00:00 | nvfp4-pack-quantized | 5 | true | 0f9ffc6a0e5620a9d6f45b330d46ab5ba1399bcd86e92f559e3c1747c9ac24d7 | 12f59baff0d71a1d | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
a2genesis/Qwen3.8-27B-NVFP4 | 2,115 | 0 | 2026-08-14T00:00:00 | modelopt | 2 | true | 84edfb86719a0d77fb125f6099a61b40e6fa25b8ed287ea5cdccfdc73e0e96ba | dc9497f6b89dc75b | 2,001 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
PassingByPixels/Qwen3.8-27B-NVFP4 | 2,069 | 2 | 2026-08-14T00:00:00 | modelopt | 149 | true | 90c4bbdb1849a41bb255b017e044184c863418fdf96bc29efd379b563a99038c | 38d5c9f5e3029194 | 2,399 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
lribeiro/Qwen3.8-27B-nvfp4-v17 | 2,041 | 0 | 2026-08-17T00:00:00 | float-quantized | 193 | true | 8d2ef045897c41455abf7e963f1799cdda2e0506571ce171e4ff03416911295a | 0baf8b06440b2673 | 1,632 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a float-quantized build, whose loader path this script does not model | null |
edp1096/Huihui-RadixArk-Qwen3.8-27B-abliterated-NVFP4 | 1,836 | 0 | 2026-08-19T00:00:00 | modelopt | 2 | true | c9ee610107d4f6e4b8eb518d23449ae6b92ddb7bf52460574468282a2ca4fa1e | 96a96d28dc135516 | 2,226 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
windowsxp811203/Qwen3.8-27B-Abliterated-NVFP4 | 1,643 | 2 | 2026-08-24T00:00:00 | nvfp4-pack-quantized | 455 | true | 84e9e65a4a8f8c510745487e35bebfd1026bccb47181feec7b91404c4f7e8455 | 877d033749c12e05 | 1,711 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
dbirks/Qwen3.8-27B-NVFP4-AutoRound | 1,600 | 3 | 2026-08-15T00:00:00 | nvfp4-pack-quantized | 215 | true | 446ac1c87957090f01b4daec158246fd3349f1c853434798c37b7e02f74f5d62 | 481545f125079395 | 2,399 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
r0b0tlab/Qwen3.8-27B-NVFP4-MTP-sm121 | 1,482 | 4 | 2026-08-16T00:00:00 | modelopt | 0 | true | 78c9be9a443c58c5c9bb516aa658d27522d16774991d112c7c321f789daa4bb6 | f6daf5b19eda2f5f | 2,001 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
Mantrah/Qwen3.8-27B-NVFP4-GDN | 1,264 | 2 | 2026-08-16T00:00:00 | modelopt | 0 | true | 66c11a87f72311b91a919af540d51c379213debc08cb9a6677450cf59d3785e7 | 3a2e18c15567d718 | 2,401 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
Vtuber-plan/Qwen3.8-27B-Uncensored-NVFP4 | 1,146 | 1 | 2026-08-20T00:00:00 | modelopt | 149 | true | fc990b87e146e1812ef1b8ccd98d15a10fcbb6073480abad9f576fbb554310ce | 38d5c9f5e3029194 | 2,399 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
windowsxp811203/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NVFP4 | 929 | 2 | 2026-08-24T00:00:00 | nvfp4-pack-quantized | 455 | true | 1f4b59f2e50ffb3b9531bcda0f4229b3d81186282eaa31d8f83e521c707a1c72 | 877d033749c12e05 | 1,711 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
Ostfralla/Qwen3.8-27B-NVFP4-NInfer | 838 | 13 | 2026-08-15T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
pottokao/Qwen3.8-27B-NVFP4-MTP-2x16GB | 741 | 2 | 2026-08-18T00:00:00 | modelopt | 1 | true | 1f530656abe23e47d69188bfcdc1caaabdd2655b7e1739bb7e40b7fc7e494b16 | bdfa6c7f2bef51b8 | 2,001 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090-LMHead4 | 725 | 1 | 2026-08-19T00:00:00 | modelopt | 148 | true | c4fc71330751f54dc06c7d6b4c5cc51bbd3005c4b518b3bb8c0fb1a6efa55349 | c2eb2f759c3ec670 | 2,402 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | duplicate-artifact | every weight shard sha256-identical to gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090, same quantization_config | gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090 |
BennyDaBall/Qwen3.8-Uncensored-NVFP4-MTP | 701 | 5 | 2026-08-21T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
vrfai/Qwen3.8-27B-NVFP4 | 688 | 3 | 2026-08-20T00:00:00 | nvfp4-pack-quantized | 303 | true | 6bdc7f66e7e984e5cf525dd7838c51466646bb8880b04e061f4587e5948a839b | fd5b050258a0a3f3 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
vroomfondel/Qwen3.8-27B-NVFP4-ModelOpt | 673 | 1 | 2026-08-15T00:00:00 | modelopt | 295 | true | 825d19443aa36f6ede5e27b73906aa20d90a319f604be043c48cae207c656301 | bae33bad5561eae7 | 2,431 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
cloudnathan5/Qwen3.8-27B-NVFP4a4-GPTQ | 665 | 0 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 447 | true | e7cddcb74f15d5017b2ce3008ea31cb576c5c35dbf48062397f2678df73029ec | 334094a88efce76d | 1,967 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | false | full | check-me | compressed-tensors build with bf16 mtp.* tensors that are absent from quantization_config.ignore; measure draft acceptance before use | null |
PassingByPixels/Qwen3.8-27B-heretic-ara-NVFP4 | 648 | 1 | 2026-08-16T00:00:00 | modelopt | 162 | true | 1ff4e05e5242fcb0114d92b3cf79a2d22e5f481b002fb33fce5dd6cac5145cc4 | b8e2a6550a63d7cb | 2,399 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
Brooooooklyn/Qwen3.8-27B-NVFP4-mlx | 640 | 1 | 2026-08-14T00:00:00 | null | 0 | true | fcab92cd2c9d8fd5cc751632c91329f3940d89bd117b052a740954eb9e5807a1 | 78fca179cf57f7f9 | 1,600 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a None build, whose loader path this script does not model | null |
lyf/Qwen3.8-27B-Huihui-Abliterated-NVFP4-MTP-VL | 636 | 1 | 2026-08-19T00:00:00 | nvfp4-pack-quantized | 479 | true | 40611dd622980f3309a665f9bf520b3aefbca51985efca071e5dbe0882798380 | 8a77c7cee1e02c77 | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
primitive-ai/Qwen3.8-27B-mixed-NVFP4-FP8 | 625 | 2 | 2026-08-24T00:00:00 | mixed-precision | 7 | true | 70c6db80975e1734627601280c67284994d786997f58d5cd77f3814e70271d46 | c93c81b71264532c | 2,151 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 312 | 144 | 336 | 0 | 312 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
philbert440/Qwen3.8-27B-Uncensored-Aggressive-NVFP4 | 603 | 2 | 2026-08-19T00:00:00 | nvfp4-pack-quantized | 407 | true | 77256816e5f43863b2401b8da230e3b7b36f2d2751eaefe0037252f05b3e84cc | 9305942a099613eb | 1,807 | 15 | 0 | 0 | 15 | 240 | 48 | 333 | 0 | 0 | 167 | 432 | 96 | 48 | 336 | 0 | 96 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
dbrasdasilva/Qwen3.8-27B-Text-NVFP4-MTP | 588 | 3 | 2026-08-15T00:00:00 | modelopt | 54 | true | f32284c7973dca3a0a36e85fb15e4f9e64b60ea7d11f13481d49d3e419ec11fd | 3f96b6ee4d44656a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
HangGlidersRule/Darkstar-Qwen3.8-27B-Abliterated-ModelOpt-W4A16-NVFP4-Mixed-FP8 | 578 | 1 | 2026-08-21T00:00:00 | modelopt | 2 | true | 19b6198078ba62bc3b082c6e70eac4ef1c1fc7d52d21bc6ac19efc64f9e530ab | b2e613060bb96e75 | 2,001 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
cloudnathan5/Qwen3.8-27B-NVFP4a4-FP8-GPTQ | 572 | 0 | 2026-08-14T00:00:00 | mixed-precision | 207 | true | 7ffc9f78e60ca9207ac9f52f0bd7f442de1e45a84921e0f176a6a5fbb243d1e0 | 11f27983a6f05d28 | 2,207 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 240 | 240 | 336 | 0 | 240 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
blkpws/Qwen3.8-27B-NVFP4-IRIS-Cache-ObjectScript | 570 | 0 | 2026-08-22T00:00:00 | mixed-precision | 97 | true | c9d06bc30e410b32ed3fad8b6c7c30f81d984791dc867d6bdd1400220b2d4e02 | b9cc15ae1e471e6d | 1,731 | 0 | 0 | 0 | 0 | 240 | 240 | 0 | 0 | 0 | 0 | 432 | 240 | 240 | 336 | 0 | 240 | false | text-only | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
zherebetskyy/Qwen3.8-27B-nvfp4-mlx | 560 | 1 | 2026-08-15T00:00:00 | null | 0 | true | e04a8b70c7c3b5396f27591f725abc304ad3d072933ca5d7ed8e2f3765d615d3 | 9c16fdf1af1c2cbf | 1,682 | 0 | 0 | 0 | 0 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 240 | 240 | 336 | 0 | 240 | false | full | duplicate-artifact | every weight shard sha256-identical to mlx-community/Qwen3.8-27B-nvfp4, same quantization_config | mlx-community/Qwen3.8-27B-nvfp4 |
TelperionAI/Qwen3.8-27B-NVFP4-AWQ-AutoRound | 547 | 5 | 2026-08-21T00:00:00 | mixed-precision | 304 | true | 8bcdb078087548fb6a828894b6a2d3c51f46dde99fdbad3af2fcdf9a477d0c9a | cc3322b65c4764ea | 1,935 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
malaiwah/Qwen3.8-27B-MTP-NVFP4-archival-6d98dc1f | 521 | 1 | 2026-08-17T00:00:00 | nvfp4-pack-quantized | 222 | true | 7384170b97da8e42de87467571142efb8b3cf560c53f7db9eaeb833d72574a11 | 1414efc09a75f72b | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | duplicate-artifact | every weight shard sha256-identical to sakamakismile/Qwen3.8-27B-MTP-NVFP4, same quantization_config | sakamakismile/Qwen3.8-27B-MTP-NVFP4 |
dfischermittwald/Qwen3.8-27B-NVFP4-DFlash2 | 491 | 1 | 2026-08-20T00:00:00 | mixed-precision | 304 | true | de035d52ca22ed29d32c30ccaf63f581b307a0826b84ca69a1d0e53cdab26a83 | d5ac492388287af9 | 1,967 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
acyildirimer/Qwen3.8-27B-NVFP4 | 443 | 1 | 2026-08-22T00:00:00 | modelopt | 2 | true | 20f681eda0769e2871b7a4674900b0cd9c3aa7aae1b4eefdd2536c91b984363d | 36cccef4ffd8dc72 | 2,001 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
Preyazz/Qwen3.8-27B-NVFP4 | 425 | 0 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 447 | true | 7fb008b391a4c2e79963db5e6da9bf6134a74ed1e27265e3e398c33f2457bbe7 | 7fdb9d39409d9b2f | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
philbert440/Qwen3.8-27B-NVFP4 | 418 | 0 | 2026-08-17T00:00:00 | nvfp4-pack-quantized | 407 | true | 51ef8e19d647d7d4343a5c6030b63bb756ffa350e1b6007ff5e66b4099f48cff | 9305942a099613eb | 1,807 | 15 | 0 | 0 | 15 | 240 | 48 | 333 | 0 | 0 | 167 | 432 | 96 | 48 | 336 | 0 | 96 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
voves/Qwen3.8-27B-NVFP4 | 407 | 0 | 2026-08-23T00:00:00 | nvfp4-pack-quantized | 207 | true | a6b2cfca702c05a0e911d86b44c3d9ca8dd6957e55dc9cecf530df4f71da1d1b | 6d5873ca518c74e2 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
philbert440/Qwen3.8-27B-Uncensored-Cyber-NVFP4 | 382 | 1 | 2026-08-19T00:00:00 | nvfp4-pack-quantized | 407 | true | 1161353c5ad1c9daab197f7f801ce8e049dcb843c96e2472f3a8aac2b9562795 | 9305942a099613eb | 1,807 | 15 | 0 | 0 | 15 | 240 | 48 | 333 | 0 | 0 | 167 | 432 | 96 | 48 | 336 | 0 | 96 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
preetpatel/Qwen3.8-27B-Uncensored-NVFP4 | 325 | 0 | 2026-08-19T00:00:00 | nvfp4-pack-quantized | 207 | true | f7b2e4a71092b88e41a54f494363bedb80f6b07b2b5334cc116f6404c3911972 | ce8f3aaeeaf39a5b | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
PreDoctor/Qwen3.8-27B-NVFP4-W4A16-PreTensor | 316 | 1 | 2026-08-24T00:00:00 | null | 0 | true | 2c7a7eebd1adc555b6767a1f8fc8d67f5a1a39525e11ef6b07414d4d662893a1 | 44136fa355b3678a | 1,986 | 0 | 0 | 0 | 0 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 480 | 240 | 336 | 0 | 480 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
zebulon-prime/Qwen3.8-27B-Dominatrix-MTP-NVFP4 | 312 | 0 | 2026-08-18T00:00:00 | modelopt | 2 | true | 470f2d35d7a0487546a17d51e28cbf47f852992114b8bc9e0c94cc15fe96efa0 | 96a96d28dc135516 | 2,194 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
lyf/Qwen3.8-27B-Huihui-Abliterated-NInfer-NVFP4 | 311 | 1 | 2026-08-19T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
ig1/Qwen3.8-27B-NVFP4-FP8KV-MTP | 289 | 0 | 2026-08-18T00:00:00 | nvfp4-pack-quantized | 448 | true | 5c98b922f3a642d4013b94a79fec4de23c51053e1005e138400108dbd2773ce1 | 7d734b921890805c | 1,999 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
tlwu/Qwen3.8-27B-NVFP4-ONNX | 276 | 0 | 2026-08-22T00:00:00 | mixed-precision | 303 | true | null | e66254de77fc4318 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
mich-pro/Qwen3.8-27B-NVFP4-Vision-MTP | 271 | 0 | 2026-08-23T00:00:00 | nvfp4-pack-quantized | 97 | true | faaf8cc658d452e20447bbdee392084a3a93b242383d19501f01f9ab385a57a3 | 154ff1b1c28bfa11 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
mmangkad/Qwen3.8-27B-NVFP4 | 245 | 0 | 2026-08-17T00:00:00 | modelopt | 2 | true | 5bab1b26c15fabf617819e7e327372ffa4fdf3c5780906577c84225dac7894a6 | b6c21d2a42579272 | 1,986 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
unistra-dnum/Qwen3.8-27B-NVFP4 | 229 | 1 | 2026-08-15T00:00:00 | nvfp4-pack-quantized | 448 | true | d78ca7e68462087571ae780302964ef0d3e675ffe6e40fa802b43bb529358406 | 6e4df2bf526ed82f | 1,967 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
kristianpaul/Qwen3.8-27B-NVFP4 | 223 | 1 | 2026-08-15T00:00:00 | modelopt | 149 | true | 72d6a99bf4812ef701817af56e23f8215b4da177275f41649cc037b7ebe5f575 | cfd3c0a5ba4f5474 | 2,431 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
YCWTG/Qwen3.8-27B-NVFP4A16-GPTQ | 214 | 0 | 2026-08-17T00:00:00 | nvfp4-pack-quantized | 447 | true | 28e5acce790f4d066b309b0bb6ed94ccf052a73f57196851a6784dcfd3618b9e | 2216f4aeba0f550e | 1,696 | 0 | 0 | 0 | 0 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
trithemius/Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-PrismQuant | 180 | 6 | 2026-08-19T00:00:00 | mixed-precision | 190 | true | c3c3a4e496f37756602acd2bb0cb50368f27051840f9cf33ea78689a8af0e37d | a8217fd8719d9c97 | 2,200 | 15 | 0 | 0 | 15 | 240 | 195 | 333 | 0 | 0 | 167 | 432 | 395 | 195 | 336 | 0 | 395 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
kekzle/Qwen3.8-27B-NVFP4-vllm | 179 | 0 | 2026-08-16T00:00:00 | nvfp4-pack-quantized | 170 | true | 81f46f09ef277e658fd742a78d76b5552143e2c41243c66105957128728c1cf7 | e097a0da4d3a1880 | 2,191 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 480 | 240 | 336 | 0 | 480 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
Peutlefaire/Qwen3.8-27B-NVFP4 | 172 | 3 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 208 | true | 68102183ab8ca02d0f4be241023ab2c69ea6a05d45830fe5b1d58d77c2196598 | 9384ceb0cd02506e | 2,687 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 720 | 240 | 336 | 0 | 720 | true | full | ok | mtp.* present, bf16, and pinned in ignore | null |
logicauro/Qwen3.8-27B-Dominatrix-NVFP4A16 | 155 | 0 | 2026-08-18T00:00:00 | nvfp4-pack-quantized | 447 | true | 9170e09a0ddbbbec9b337ef37b0fda7a754471845f30920b7150fe18b8020714 | f2b2ba4d0cec6ddf | 1,696 | 0 | 0 | 0 | 0 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
lribeiro/Qwen3.8-27B-nvfp4-v6 | 149 | 0 | 2026-08-16T00:00:00 | mixed-precision | 192 | true | bdb7595347731ee9bd19a6ae983400829ddef41d2c4134a19b3f81acddd9c8b3 | 564720301bae3563 | 1,380 | 0 | 0 | 0 | 0 | 240 | 144 | 0 | 0 | 0 | 0 | 432 | 144 | 144 | 336 | 0 | 144 | false | text-only | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
t0rr3sp3dr0/Qwen3.8-27B-MLX-MTP-nvfp4 | 145 | 0 | 2026-08-19T00:00:00 | null | 0 | true | 8cdb18495b9e96186475b403e78f88fce9b0f3625c397401c18ad38fa26f8d42 | 9c16fdf1af1c2cbf | 1,703 | 15 | 7 | 7 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 240 | 240 | 336 | 7 | 240 | false | full | mtp-quantized | all 15 logical mtp.* tensors present but quantized (7 scale keys); verify acceptance rate before trusting | null |
reality-interface/qwen3.8-27b-agentic-nvfp4 | 141 | 1 | 2026-08-23T00:00:00 | modelopt | 148 | true | c4fc71330751f54dc06c7d6b4c5cc51bbd3005c4b518b3bb8c0fb1a6efa55349 | c2eb2f759c3ec670 | 2,402 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | duplicate-artifact | every weight shard sha256-identical to gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090, same quantization_config | gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090 |
kekzle/Qwen3.8-27B-NVFP4 | 139 | 1 | 2026-08-16T00:00:00 | null | 0 | true | b8eac4607cd96092b8c36d98e0fc9d54aee5f35c32a53032f1e0a947f6ae1912 | 44136fa355b3678a | 2,191 | 15 | 0 | 0 | 15 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 480 | 240 | 336 | 0 | 480 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a None build, whose loader path this script does not model | null |
TelperionAI/Qwen3.8-27B-NVFP4-AWQ-GPTQ | 136 | 3 | 2026-08-21T00:00:00 | mixed-precision | 304 | true | 64e96aa96e30533155def017fd241d7ad2eb4b81e9c09b857a53c9bbe83d65ec | 7c45add40de1cc46 | 1,935 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
eemin/Qwen3.8-27B-NVFP4-MTP-2x16GB | 131 | 0 | 2026-08-19T00:00:00 | modelopt | 1 | true | 1f530656abe23e47d69188bfcdc1caaabdd2655b7e1739bb7e40b7fc7e494b16 | bdfa6c7f2bef51b8 | 2,001 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | duplicate-artifact | every weight shard sha256-identical to pottokao/Qwen3.8-27B-NVFP4-MTP-2x16GB, same quantization_config | pottokao/Qwen3.8-27B-NVFP4-MTP-2x16GB |
AudioJaguar/Qwen3.8-27B-Uncensored-NVFP4-W4A4 | 123 | 0 | 2026-08-23T00:00:00 | modelopt | 2 | true | 499eea1c0d3580e6ef4f720d2c8862905b6c74b3f231df767f6b26265a0b35dd | 1512aa69d3dcb019 | 2,194 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
johnnyeric/Qwen3.8-27B-Uncensored-NVFP4 | 103 | 0 | 2026-08-20T00:00:00 | mixed-precision | 303 | true | e8a5d7aad486c886b8089dd538d30ede12aeef9a206650f2ac772d081cc22318 | 64deb9ea9aff8f2f | 1,967 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
dsikka/Qwen3.8-27B-NVFP4-FP8-GPTQ-AWQ | 102 | 0 | 2026-08-17T00:00:00 | mixed-precision | 302 | true | bc293c5d847d61a41ee9dc0725bbdd4c3e9d2c4853dd9b5e6f36f31b3a2d7807 | c0aee00f5fe1a788 | 1,953 | 0 | 0 | 0 | 0 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
cloudnathan5/Qwen3.8-27B-NVFP4a16-GPTQ | 101 | 0 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 447 | true | f7d46630782c6328db9d8eb4c513ddb0723a5870873869ac29cd6dd330c9278f | 2216f4aeba0f550e | 1,711 | 15 | 0 | 0 | 15 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | false | full | check-me | compressed-tensors build with bf16 mtp.* tensors that are absent from quantization_config.ignore; measure draft acceptance before use | null |
berkerdooo/Qwen3.8-27B-NVFP4-1M | 96 | 0 | 2026-08-20T00:00:00 | mixed-precision | 302 | true | 4ef52f8f2ccb61f5164e49f02df8c31f5c057e5fb658d507ec742e7980e66de3 | bc416b2853d3f51a | 1,968 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | false | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
dudeman2512/Qwen3.8-27B-NVFP4 | 92 | 0 | 2026-08-18T00:00:00 | nvfp4-pack-quantized | 7 | true | 0157be4de05c5d6e265f6d25d87f77e1a47f76bdea2ee03e06ea52f6f1778d06 | 8dad7d9e95452095 | 2,429 | 15 | 16 | 8 | 15 | 240 | 240 | 333 | 222 | 111 | 167 | 432 | 480 | 240 | 336 | 16 | 480 | false | full | mtp-quantized | all 15 logical mtp.* tensors present but quantized (16 scale keys); verify acceptance rate before trusting | null |
rzgar/Qwen3.8-27B-NVFP4-ComfyUI | 86 | 4 | 2026-08-15T00:00:00 | null | 0 | true | 0b917f8a3d40fda7eb2224444fde5e6a52fe848e7c2b93f6070553e6b1072639 | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
mikelalda/qwen3.8-27b_euskara-NVFP4 | 83 | 0 | 2026-08-20T00:00:00 | mixed-precision | 303 | true | a481ca0ac03e11b0343d76b8c499f3ff5cb5c0079870ba0d725b756250bb3bfe | 8f9d47c02839f442 | 1,968 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
gearwave00001/Huihui-Qwen3.8-27B-abliterated-nvfp4-NInfer | 75 | 2 | 2026-08-19T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
JasonW2025/Qwen3.8-27B-NVFP4-W4A16 | 63 | 0 | 2026-08-21T00:00:00 | modelopt | 2 | true | 24eb1fa5a44b6c8938edf0396d489b09c1c1191c889ab2756c0766eeecd5803d | 4ac98ba08ef35394 | 2,033 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 288 | 144 | 336 | 0 | 288 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a modelopt build, whose loader path this script does not model | null |
ThakiCloud/Qwen3.8-27B-NVFP4-FP8ATTN | 62 | 1 | 2026-08-23T00:00:00 | mixed-precision | 207 | true | 744dcc83f698deb4824ff60e17fc0cf9b49e3f0e40ae3ab1414f3e6967d965f7 | b9f5a4d714803875 | 2,064 | 0 | 0 | 0 | 0 | 240 | 240 | 333 | 0 | 0 | 167 | 432 | 240 | 240 | 336 | 0 | 240 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
nm-testing/Qwen3.8-27B-NVFP4-FP8 | 60 | 0 | 2026-08-16T00:00:00 | mixed-precision | 303 | true | 7bb0ee3a8295a33cba2fdfcfe39cfd5f5be26d902be5cb8b96f7d278ff7abef8 | ce9a2954c8aa1972 | 1,968 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
gittensor-model-hub/Qwen3.8-27B-NVFP4-RTX5090-No-MTP | 58 | 0 | 2026-08-19T00:00:00 | modelopt | 149 | false | fec674660caa889ad05beb3a11febc4a16c662a33d1ac3db40b5946358b5c8a4 | 6040c1e55e6a98c1 | 2,384 | 0 | 0 | 0 | 0 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 432 | 144 | 336 | 0 | 432 | true | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it | null |
cometkim/Qwen3.8-27B-nvfp4full-NInfer | 57 | 1 | 2026-08-20T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
kyaky/Qwen3.8-27B-NVFP4 | 56 | 0 | 2026-08-20T00:00:00 | mixed-precision | 306 | true | 1421ec3040e1add64dbc63a985c2b2c67f6589de203694156123e315387aed27 | 81c961feb8ab1e48 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
YourHighnessLA/Qwen3.8-27B-DFlash2-NVFP4 | 54 | 0 | 2026-08-22T00:00:00 | nvfp4-pack-quantized | 3 | false | a0cc8980c685bd92fa57178e72b03ce008f5cd0f98619e6f0625c28fab77ec18 | 279b2874a088d6c2 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
TelperionAI/Huihui-Qwen3.8-27B-abliterated-NVFP4 | 54 | 0 | 2026-08-21T00:00:00 | mixed-precision | 304 | true | 36ca31306b3194a6bcd775762a5d84bac2420332f513e23b55d141e552436639 | f17b354bfb6ba5e9 | 1,935 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
gearwave00001/orcarouter-Qwen3.8-27B-Uncensored-nvfp4-NInfer | 51 | 0 | 2026-08-20T00:00:00 | null | 0 | false | null | 44136fa355b3678a | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | not-auditable | no safetensors index (single-file checkpoint, gated repo, or non-safetensors layout) | null |
ThakiCloud/Qwen3.8-27B-NVFP4-GPTQ-txt | 48 | 0 | 2026-08-23T00:00:00 | nvfp4-pack-quantized | 511 | true | ee5074b4654306fe7e0fa64efd23561d41b3be743be8aa11a7525d93ae8118b3 | 1a73be00ed4028e3 | 1,760 | 0 | 0 | 0 | 0 | 240 | 0 | 333 | 0 | 0 | 167 | 432 | 0 | 0 | 336 | 0 | 0 | false | full | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
HivenetQuant/Qwen3.8-27B-NVFP4 | 47 | 3 | 2026-08-21T00:00:00 | mixed-precision | 313 | true | 7f9d31896472de68057b14bb76f368ae4e8ec3766453f1014426c7103eac7bbc | 8c55225318c21177 | 2,099 | 15 | 0 | 0 | 15 | 240 | 144 | 333 | 0 | 0 | 167 | 432 | 144 | 144 | 336 | 0 | 144 | true | full | ok-unverified-schema | mtp.* present and unquantized, but this is a mixed-precision build, whose loader path this script does not model | null |
lokeshe09/Qwen3.8-27B-NVFP4 | 47 | 0 | 2026-08-14T00:00:00 | nvfp4-pack-quantized | 337 | true | 5320ccd68ec0d3ea5f04c502da270bda207f673c85489105cd5b0d2ffbfa33d6 | 0c868793eb35d281 | 1,619 | 0 | 0 | 0 | 0 | 240 | 0 | 0 | 0 | 0 | 0 | 432 | 0 | 0 | 336 | 0 | 0 | false | text-only | no-mtp-head | checkpoint contains no mtp.* tensors, so vLLM cannot run --speculative-config method=mtp against it; worse, config.json still declares an MTP layer count, promising a head that isn't there | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.