From 7d4f6e1a064d443b2c78d98485596c17857e6f97 Mon Sep 17 00:00:00 2001 From: ModelHub XC Date: Sat, 30 May 2026 12:11:17 +0800 Subject: [PATCH] =?UTF-8?q?=E5=88=9D=E5=A7=8B=E5=8C=96=E9=A1=B9=E7=9B=AE?= =?UTF-8?q?=EF=BC=8C=E7=94=B1ModelHub=20XC=E7=A4=BE=E5=8C=BA=E6=8F=90?= =?UTF-8?q?=E4=BE=9B=E6=A8=A1=E5=9E=8B?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Model: ZuzeTt/Qwen3-VL-4B-Instruct-heretic-Imatrix-GGUF Source: Original Platform --- .gitattributes | 71 ++++++ Qwen3-VL-4B-Instract-heretic-f16.gguf | 3 + ...-VL-4B-Instract-heretic-imatrix-IQ1_M.gguf | 3 + ...-VL-4B-Instract-heretic-imatrix-IQ1_S.gguf | 3 + ...-VL-4B-Instract-heretic-imatrix-IQ2_S.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-IQ2_XS.gguf | 3 + ...L-4B-Instract-heretic-imatrix-IQ2_XXS.gguf | 3 + ...-VL-4B-Instract-heretic-imatrix-IQ3_S.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-IQ3_XS.gguf | 3 + ...L-4B-Instract-heretic-imatrix-IQ3_XXS.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-IQ4_NL.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-IQ4_XS.gguf | 3 + ...3-VL-4B-Instract-heretic-imatrix-Q2_K.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q3_K_L.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q3_K_M.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q3_K_S.gguf | 3 + ...3-VL-4B-Instract-heretic-imatrix-Q4_0.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q4_K_M.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q4_K_S.gguf | 3 + ...3-VL-4B-Instract-heretic-imatrix-Q5_0.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q5_K_M.gguf | 3 + ...VL-4B-Instract-heretic-imatrix-Q5_K_S.gguf | 3 + ...3-VL-4B-Instract-heretic-imatrix-Q6_K.gguf | 3 + ...3-VL-4B-Instract-heretic-imatrix-Q8_0.gguf | 3 + Qwen3-VL-4B-Instract-heretic-mmproj-f16.gguf | 3 + README.md | 224 ++++++++++++++++++ 26 files changed, 367 insertions(+) create mode 100644 .gitattributes create mode 100644 Qwen3-VL-4B-Instract-heretic-f16.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_M.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_S.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_S.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XS.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XXS.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_S.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XS.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XXS.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_NL.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_XS.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q2_K.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_L.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_M.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_S.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q4_0.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_M.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_S.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q5_0.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_M.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_S.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q6_K.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-imatrix-Q8_0.gguf create mode 100644 Qwen3-VL-4B-Instract-heretic-mmproj-f16.gguf create mode 100644 README.md diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..2447539 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,71 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-f16.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-mmproj-f16.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q4_0-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q4_K_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q8_0-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q4_K_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q2_K-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q3_K_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q3_K_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q5_0-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q5_K_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q5_K_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q6_K-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q4_K_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q4_K_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q2_K-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q3_K_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q4_0-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q6_K-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q3_K_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q5_K_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-f16.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q5_K_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q8_0-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-4B-Thinking-2507-Heretic-Q5_0-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ1_M-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ1_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ2_XS-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ2_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ2_XXS-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ3_S-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ3_XS-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ3_XXS-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-Q3_K_L-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ4_NL-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +Qwen3-VL-4B-Instract-heretic-IQ4_XS-imatrix.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/Qwen3-VL-4B-Instract-heretic-f16.gguf b/Qwen3-VL-4B-Instract-heretic-f16.gguf new file mode 100644 index 0000000..bf06897 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-f16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4eba5b6f8059318d903b7e46ea78a446739d14e06fbd87ca2801457ae0b6404 +size 8051286368 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_M.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_M.gguf new file mode 100644 index 0000000..dcfa651 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c724809fa25ef560f5789765353b83aa122f7dfb50e60e1b91c4c939adfbdb1f +size 1127019104 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_S.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_S.gguf new file mode 100644 index 0000000..e524831 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ1_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f27bf1528cb40b29eff58d59637201f0e743978f423206e77f9edd1eba65e9f1 +size 1055257184 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_S.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_S.gguf new file mode 100644 index 0000000..1fcb5cd --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ccf16e99659c98e2197bd5bfef19fe6cfa35148dda21fbbdf6e78efd9bb35a34 +size 1417302624 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XS.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XS.gguf new file mode 100644 index 0000000..2ef7de0 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0471acfef7d3e9402c7a30a29a6b04590f296131afea621418432ba70dbede1 +size 1354101344 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XXS.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XXS.gguf new file mode 100644 index 0000000..8f91d5d --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ2_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b57c7b10b59aa0cc6c2278d8923881a7af017f71f748a8e58040524dbdd43dc +size 1246622304 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_S.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_S.gguf new file mode 100644 index 0000000..9f60c64 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0dc54e33d61b84aa8ba67ecd6def1e46ea25ac1cdf9ba13e7d9f33b766325d73 +size 1899532384 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XS.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XS.gguf new file mode 100644 index 0000000..3387c65 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:386d2e3b1e360cc224a63d1f63d6323afec5be9e355d9bb45697c58403a6ec22 +size 1814376544 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XXS.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XXS.gguf new file mode 100644 index 0000000..17166cb --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db82b58c9433c779a0611033d048c8692c9cf77bae6b62e8828ed1a79082935b +size 1670189664 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_NL.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_NL.gguf new file mode 100644 index 0000000..53540d5 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9edda58af4b2cde664ecdb9d53ee15c55649d15e13b1c164ef59a1ce1e3df101 +size 2381344864 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_XS.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_XS.gguf new file mode 100644 index 0000000..4d8c0d8 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b23ed9492c841d350171a54893f8a086e640902919b755da0af4a904faf7dbe1 +size 2270752864 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q2_K.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q2_K.gguf new file mode 100644 index 0000000..1ed9c1f --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:34a52431759ea614d0fd4109a9dcba5fb7f0679162157f046268d22e8c005d26 +size 1669501024 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_L.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_L.gguf new file mode 100644 index 0000000..9fc048d --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b76881b24bcb682e1095b70bb643aed7b0cbec1707654cf20588507b91892c9 +size 2239787104 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_M.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_M.gguf new file mode 100644 index 0000000..1c9d693 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:990194f0068a05cea362a48a5f784e11cc85f928115f368d1ced4c8c990af75b +size 2075619424 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_S.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_S.gguf new file mode 100644 index 0000000..0088489 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4fa9682c9722cc0a2c0c22ff7576133d811a46269a0c2e4f405f047c312bd82f +size 1886998624 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_0.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_0.gguf new file mode 100644 index 0000000..f82172d --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7a80acfd072e20a62b2fc30e1792d4629afbc3eeaf3a7d96701932e55a23d89 +size 2375774304 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_M.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_M.gguf new file mode 100644 index 0000000..77ad238 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:73e76145456ef3b0f562e575681b0d4d05356c42b5de86f318bc5315b4d6e0ac +size 2497282144 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_S.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_S.gguf new file mode 100644 index 0000000..fa145f4 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09e9aba8afd768943bb333efc0189f4ab649e49c9eecfb3be1d878c3e68e4086 +size 2383310944 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_0.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_0.gguf new file mode 100644 index 0000000..7872131 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf4993f1d2672f40406f3560b651c400dc8aeb604a02247cdc1d560c0eb60ab1 +size 2829938784 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_M.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_M.gguf new file mode 100644 index 0000000..ab84cc6 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7370ac5825473620f6278908b12ab01ab69bf89b3f8ac99bb5b28be4fb01c85d +size 2889515104 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_S.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_S.gguf new file mode 100644 index 0000000..46c2d64 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84ff9952883eb1d5268976a628b156dc57173aa9dc6776898266d09927642390 +size 2823712864 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q6_K.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q6_K.gguf new file mode 100644 index 0000000..97467f3 --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:afdbbf137187f0016a23ce7888718be5e537290d9d020f9bca23e70a8224d3fd +size 3306262624 diff --git a/Qwen3-VL-4B-Instract-heretic-imatrix-Q8_0.gguf b/Qwen3-VL-4B-Instract-heretic-imatrix-Q8_0.gguf new file mode 100644 index 0000000..63c67ed --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-imatrix-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:991e4cc2c46e4fec1d29eb2ad0de9d1a87058fc1961d6adffe03b1187375a16c +size 4280406624 diff --git a/Qwen3-VL-4B-Instract-heretic-mmproj-f16.gguf b/Qwen3-VL-4B-Instract-heretic-mmproj-f16.gguf new file mode 100644 index 0000000..d5b751b --- /dev/null +++ b/Qwen3-VL-4B-Instract-heretic-mmproj-f16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d20fea7b0226a265889448d6c52d2ebe3c8604a7ff77d3e2088f0e4a588ffed +size 836180448 diff --git a/README.md b/README.md new file mode 100644 index 0000000..a930864 --- /dev/null +++ b/README.md @@ -0,0 +1,224 @@ +--- +license: apache-2.0 +pipeline_tag: image-text-to-text +library_name: transformers +tags: +- heretic +- uncensored +- decensored +- abliterated +--- +# This is a decensored version of [Qwen/Qwen3-VL-4B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct), made using [Heretic](https://github.com/p-e-w/heretic) v1.0.1 + +Original model: https://huggingface.co/georgehenney/Qwen3-VL-4B-Instruct-heretic-8refusal + +## Abliteration parameters + +| Parameter | Value | +| :-------- | :---: | +| **direction_index** | 15.17 | +| **attn.o_proj.max_weight** | 1.47 | +| **attn.o_proj.max_weight_position** | 25.63 | +| **attn.o_proj.min_weight** | 1.07 | +| **attn.o_proj.min_weight_distance** | 18.22 | +| **mlp.down_proj.max_weight** | 1.17 | +| **mlp.down_proj.max_weight_position** | 27.14 | +| **mlp.down_proj.min_weight** | 1.07 | +| **mlp.down_proj.min_weight_distance** | 19.55 | + +## Performance + +| Metric | This model | Original model ([Qwen/Qwen3-VL-4B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct)) | +| :----- | :--------: | :---------------------------: | +| **KL divergence** | 0.27 | 0 *(by definition)* | +| **Refusals** | 8/100 | 92/100 | + +----- + + + Chat + + + +# Qwen3-VL-4B-Instruct + + +Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date. + +This generation delivers comprehensive upgrades across the board: superior text understanding & generation, deeper visual perception & reasoning, extended context length, enhanced spatial and video dynamics comprehension, and stronger agent interaction capabilities. + +Available in Dense and MoE architectures that scale from edge to cloud, with Instruct and reasoning‑enhanced Thinking editions for flexible, on‑demand deployment. + + +#### Key Enhancements: + +* **Visual Agent**: Operates PC/mobile GUIs—recognizes elements, understands functions, invokes tools, completes tasks. + +* **Visual Coding Boost**: Generates Draw.io/HTML/CSS/JS from images/videos. + +* **Advanced Spatial Perception**: Judges object positions, viewpoints, and occlusions; provides stronger 2D grounding and enables 3D grounding for spatial reasoning and embodied AI. + +* **Long Context & Video Understanding**: Native 256K context, expandable to 1M; handles books and hours-long video with full recall and second-level indexing. + +* **Enhanced Multimodal Reasoning**: Excels in STEM/Math—causal analysis and logical, evidence-based answers. + +* **Upgraded Visual Recognition**: Broader, higher-quality pretraining is able to “recognize everything”—celebrities, anime, products, landmarks, flora/fauna, etc. + +* **Expanded OCR**: Supports 32 languages (up from 19); robust in low light, blur, and tilt; better with rare/ancient characters and jargon; improved long-document structure parsing. + +* **Text Understanding on par with pure LLMs**: Seamless text–vision fusion for lossless, unified comprehension. + + +#### Model Architecture Updates: + +

+ +

+ + +1. **Interleaved-MRoPE**: Full‑frequency allocation over time, width, and height via robust positional embeddings, enhancing long‑horizon video reasoning. + +2. **DeepStack**: Fuses multi‑level ViT features to capture fine‑grained details and sharpen image–text alignment. + +3. **Text–Timestamp Alignment:** Moves beyond T‑RoPE to precise, timestamp‑grounded event localization for stronger video temporal modeling. + +This is the weight repository for Qwen3-VL-4B-Instruct. + + +--- + +## Model Performance + +**Multimodal performance** + +![](https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-VL/qwen3vl_4b_8b_vl_instruct.jpg) + +**Pure text performance** +![](https://qianwen-res.oss-accelerate.aliyuncs.com/Qwen3-VL/qwen3vl_4b_8b_text_instruct.jpg) + +## Quickstart + +Below, we provide simple examples to show how to use Qwen3-VL with 🤖 ModelScope and 🤗 Transformers. + +The code of Qwen3-VL has been in the latest Hugging Face transformers and we advise you to build from source with command: +``` +pip install git+https://github.com/huggingface/transformers +# pip install transformers==4.57.0 # currently, V4.57.0 is not released +``` + +### Using 🤗 Transformers to Chat + +Here we show a code snippet to show how to use the chat model with `transformers`: + +```python +from transformers import Qwen3VLForConditionalGeneration, AutoProcessor + +# default: Load the model on the available device(s) +model = Qwen3VLForConditionalGeneration.from_pretrained( + "Qwen/Qwen3-VL-4B-Instruct", dtype="auto", device_map="auto" +) + +# We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios. +# model = Qwen3VLForConditionalGeneration.from_pretrained( +# "Qwen/Qwen3-VL-4B-Instruct", +# dtype=torch.bfloat16, +# attn_implementation="flash_attention_2", +# device_map="auto", +# ) + +processor = AutoProcessor.from_pretrained("Qwen/Qwen3-VL-4B-Instruct") + +messages = [ + { + "role": "user", + "content": [ + { + "type": "image", + "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg", + }, + {"type": "text", "text": "Describe this image."}, + ], + } +] + +# Preparation for inference +inputs = processor.apply_chat_template( + messages, + tokenize=True, + add_generation_prompt=True, + return_dict=True, + return_tensors="pt" +) +inputs = inputs.to(model.device) + +# Inference: Generation of the output +generated_ids = model.generate(**inputs, max_new_tokens=128) +generated_ids_trimmed = [ + out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids) +] +output_text = processor.batch_decode( + generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False +) +print(output_text) +``` + +### Generation Hyperparameters +#### VL +```bash +export greedy='false' +export top_p=0.8 +export top_k=20 +export temperature=0.7 +export repetition_penalty=1.0 +export presence_penalty=1.5 +export out_seq_length=16384 +``` + +#### Text +```bash +export greedy='false' +export top_p=1.0 +export top_k=40 +export repetition_penalty=1.0 +export presence_penalty=2.0 +export temperature=1.0 +export out_seq_length=32768 +``` + + +## Citation + +If you find our work helpful, feel free to give us a cite. + +``` +@misc{qwen3technicalreport, + title={Qwen3 Technical Report}, + author={Qwen Team}, + year={2025}, + eprint={2505.09388}, + archivePrefix={arXiv}, + primaryClass={cs.CL}, + url={https://arxiv.org/abs/2505.09388}, +} + +@article{Qwen2.5-VL, + title={Qwen2.5-VL Technical Report}, + author={Bai, Shuai and Chen, Keqin and Liu, Xuejing and Wang, Jialin and Ge, Wenbin and Song, Sibo and Dang, Kai and Wang, Peng and Wang, Shijie and Tang, Jun and Zhong, Humen and Zhu, Yuanzhi and Yang, Mingkun and Li, Zhaohai and Wan, Jianqiang and Wang, Pengfei and Ding, Wei and Fu, Zheren and Xu, Yiheng and Ye, Jiabo and Zhang, Xi and Xie, Tianbao and Cheng, Zesen and Zhang, Hang and Yang, Zhibo and Xu, Haiyang and Lin, Junyang}, + journal={arXiv preprint arXiv:2502.13923}, + year={2025} +} + +@article{Qwen2VL, + title={Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution}, + author={Wang, Peng and Bai, Shuai and Tan, Sinan and Wang, Shijie and Fan, Zhihao and Bai, Jinze and Chen, Keqin and Liu, Xuejing and Wang, Jialin and Ge, Wenbin and Fan, Yang and Dang, Kai and Du, Mengfei and Ren, Xuancheng and Men, Rui and Liu, Dayiheng and Zhou, Chang and Zhou, Jingren and Lin, Junyang}, + journal={arXiv preprint arXiv:2409.12191}, + year={2024} +} + +@article{Qwen-VL, + title={Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond}, + author={Bai, Jinze and Bai, Shuai and Yang, Shusheng and Wang, Shijie and Tan, Sinan and Wang, Peng and Lin, Junyang and Zhou, Chang and Zhou, Jingren}, + journal={arXiv preprint arXiv:2308.12966}, + year={2023} +} +``` \ No newline at end of file