tt-hous commited on
Commit
97e3989
·
verified ·
1 Parent(s): cd87682

tt-model push embeddinggemma-2-p150 (container)

Browse files
Files changed (32) hide show
  1. .gitattributes +14 -0
  2. README.md +6 -7
  3. image/blobs/sha256/0bd59b87edea4b92e96656bdead06a361eac00643fecbca5a1a247c55a460fed +28 -0
  4. image/blobs/sha256/1bfe1becf29e080cde867ea83273cc271f89d4aa5d1e03072d3d4c2023fd1ad5 +0 -0
  5. image/blobs/sha256/229487199676033e7f4475017c0d8717d719429ce5eeae256e90dd4160c2925f +3 -0
  6. image/blobs/sha256/246bf69d316f37f893fdd6ca0e1b6a9dfd2496df809d934747e89f0004dfdfd0 +0 -0
  7. image/blobs/sha256/3129b0782917119f77fef8dbd067b1bc25dc62520c9b4e0f79d194e33916637a +3 -0
  8. image/blobs/sha256/3634bcd964bae60bc52e5812c0ba3d526ff81f0973e0e9aab54aa7bb32a9075d +3 -0
  9. image/blobs/sha256/398b141b0c8b4ff904b74e545fd23e595a51b12a60a34ffa647311c82b65acf0 +19 -0
  10. image/blobs/sha256/3e88ace052c0b2d77a8aec33330ed2387af4a0d4f70ee382ece9faa26739a799 +0 -0
  11. image/blobs/sha256/44a10ad77f89c27c6dcf8a4489fe0c2cd4bc209d263ccb9c1f3350c5683cf635 +3 -0
  12. image/blobs/sha256/468f5474fec5781afd1d4819ed518b57789ecddf1a33c4b26e4ae24b943beb2b +1 -0
  13. image/blobs/sha256/4b30923806c45caff71513ec3e81797253f5bc240d27874612b4bbccae95efad +3 -0
  14. image/blobs/sha256/4bbd042d5a48f78d42374a9ff7099c61ef4155fce27d055a3e5d572ea2887230 +0 -0
  15. image/blobs/sha256/510954577bc5f9942a88171596d66e859741843fbe9eaac1f58e6268baf88c5e +0 -0
  16. image/blobs/sha256/5a009e1245fa4110a9b1e473cd2feba0a201d8d85dbec0941269e4592d27c506 +3 -0
  17. image/blobs/sha256/5d85ec83334f0a3af355a319fbf5bbf743180ace7d77fe9081a84a78cc2f3f3d +3 -0
  18. image/blobs/sha256/65de262429bc470f1e6c7de03c08d57a9eb77a30ba82d226fecce3aafbeb306c +3 -0
  19. image/blobs/sha256/668c29e85693f432d66ce28da65b9880d72f8152431dc87462f2155aa0d29105 +1 -0
  20. image/blobs/sha256/9dd63fc5eb57989e0662b806e787ab5f6f1be5072b780ef84ae960c9aaf27439 +1 -0
  21. image/blobs/sha256/b349c5f82ec7e1c9285f4c41a294f1aa9efebbd757791dc3712f4b869d2b4cd5 +3 -0
  22. image/blobs/sha256/b7aba87d343d132afe8e86260c91474534ceabd8928b068eff7e3669a861b3ec +126 -0
  23. image/blobs/sha256/b891528e3c01140e7ab06a4836d79cff05712ae577e3b798c81b5de12095d35b +3 -0
  24. image/blobs/sha256/be74ff0f8fffc8c250798344d8e2fa17f6a6a69c79586273b64b1fcf4b55c713 +3 -0
  25. image/blobs/sha256/d7161b2ea02a42b9f3b038066bb83ede992815f0461a5e35644ca91594aa6ac6 +3 -0
  26. image/blobs/sha256/d853bc77781c2b20971eb04b2a3bd79e993f899c6ac962b5211ba7bf81064a22 +0 -0
  27. image/blobs/sha256/df20c74836f348e00350c8f20fc9ec5948ed47ebdaaa0321afa85de40d341fab +0 -0
  28. image/blobs/sha256/e10f09843c9b4d73e9fd19f406e6bd9634c867ccdc92552898cfc760fc56ef6f +3 -0
  29. image/blobs/sha256/f38ef5e3e335490fe8905045047f1bd8276adc0add86089deceb7151d573dd42 +3 -0
  30. image/index.json +1 -1
  31. image/manifest.json +1 -1
  32. tt_kernel_manifest.json +9 -9
.gitattributes CHANGED
@@ -100,3 +100,17 @@ image/blobs/sha256/b65b506905faa6bdd4125549d97257e64c8a5715c11c9780b6a9b7678fd05
100
  image/blobs/sha256/c71cc42b78a54d49663258ad776fe0ceb6402a360b1b699e6d709464dfec3597 filter=lfs diff=lfs merge=lfs -text
101
  image/blobs/sha256/cf1a5487baaf7d71eb78fd0888933c6cdecb40e59edb896511aa0ef4a896031e filter=lfs diff=lfs merge=lfs -text
102
  image/blobs/sha256/fa71074b04af7e9df6466c49c50e461c6f6e1d986900d94092fbb3994a598a53 filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
100
  image/blobs/sha256/c71cc42b78a54d49663258ad776fe0ceb6402a360b1b699e6d709464dfec3597 filter=lfs diff=lfs merge=lfs -text
101
  image/blobs/sha256/cf1a5487baaf7d71eb78fd0888933c6cdecb40e59edb896511aa0ef4a896031e filter=lfs diff=lfs merge=lfs -text
102
  image/blobs/sha256/fa71074b04af7e9df6466c49c50e461c6f6e1d986900d94092fbb3994a598a53 filter=lfs diff=lfs merge=lfs -text
103
+ image/blobs/sha256/229487199676033e7f4475017c0d8717d719429ce5eeae256e90dd4160c2925f filter=lfs diff=lfs merge=lfs -text
104
+ image/blobs/sha256/3129b0782917119f77fef8dbd067b1bc25dc62520c9b4e0f79d194e33916637a filter=lfs diff=lfs merge=lfs -text
105
+ image/blobs/sha256/3634bcd964bae60bc52e5812c0ba3d526ff81f0973e0e9aab54aa7bb32a9075d filter=lfs diff=lfs merge=lfs -text
106
+ image/blobs/sha256/44a10ad77f89c27c6dcf8a4489fe0c2cd4bc209d263ccb9c1f3350c5683cf635 filter=lfs diff=lfs merge=lfs -text
107
+ image/blobs/sha256/4b30923806c45caff71513ec3e81797253f5bc240d27874612b4bbccae95efad filter=lfs diff=lfs merge=lfs -text
108
+ image/blobs/sha256/5a009e1245fa4110a9b1e473cd2feba0a201d8d85dbec0941269e4592d27c506 filter=lfs diff=lfs merge=lfs -text
109
+ image/blobs/sha256/5d85ec83334f0a3af355a319fbf5bbf743180ace7d77fe9081a84a78cc2f3f3d filter=lfs diff=lfs merge=lfs -text
110
+ image/blobs/sha256/65de262429bc470f1e6c7de03c08d57a9eb77a30ba82d226fecce3aafbeb306c filter=lfs diff=lfs merge=lfs -text
111
+ image/blobs/sha256/b349c5f82ec7e1c9285f4c41a294f1aa9efebbd757791dc3712f4b869d2b4cd5 filter=lfs diff=lfs merge=lfs -text
112
+ image/blobs/sha256/b891528e3c01140e7ab06a4836d79cff05712ae577e3b798c81b5de12095d35b filter=lfs diff=lfs merge=lfs -text
113
+ image/blobs/sha256/be74ff0f8fffc8c250798344d8e2fa17f6a6a69c79586273b64b1fcf4b55c713 filter=lfs diff=lfs merge=lfs -text
114
+ image/blobs/sha256/d7161b2ea02a42b9f3b038066bb83ede992815f0461a5e35644ca91594aa6ac6 filter=lfs diff=lfs merge=lfs -text
115
+ image/blobs/sha256/e10f09843c9b4d73e9fd19f406e6bd9634c867ccdc92552898cfc760fc56ef6f filter=lfs diff=lfs merge=lfs -text
116
+ image/blobs/sha256/f38ef5e3e335490fe8905045047f1bd8276adc0add86089deceb7151d573dd42 filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -4,7 +4,6 @@ tags:
4
  - p150
5
  - tt-dit-server
6
  - tt-model-cache
7
- - tt-model-catalog
8
  - tt-model-container
9
  license: apache-2.0
10
  pipeline_tag: feature-extraction
@@ -46,12 +45,12 @@ Embeddings for retrieval, semantic search, clustering and classification over te
46
  ## Quickstart
47
 
48
  ```bash
49
- tt serve tt-hous/embeddinggemma-2-p150
50
  ```
51
 
52
  `tt serve` downloads the Docker image and the [`google/embeddinggemma-2`](https://huggingface.co/google/embeddinggemma-2) at `914f7f89142e33e77833254d9c9b90c3cef7303b` weights (into your HF cache; they are not in the image), then starts the model's own HTTP server on port 20000 (or the next free port, if that one is busy). The first start compiles kernels for your device, which takes several minutes; the server is ready when it logs `Application startup complete`.
53
 
54
- Without tt-cli: `tt-model serve tt-hous/embeddinggemma-2-p150`.
55
 
56
  ### Embedding text
57
  ```bash
@@ -61,7 +60,7 @@ curl -s localhost:20000/v1/embeddings -H 'content-type: application/json' \
61
 
62
  ### Demo
63
  ```bash
64
- tt serve tt-hous/embeddinggemma-2-p150 --profile demo
65
  ```
66
  Open `http://<host>:20000/demo/`. A web page where every query is embedded live on the chip, over a bundled library of 163 openly licensed images, videos, sounds, texts in 10 languages and code snippets: media search, finding a moment in a long video, audio search, code search, multilingual search, and a side-by-side Matryoshka comparison.
67
 
@@ -86,7 +85,7 @@ This package exposes the model's own HTTP server on port 20000, **not** an OpenA
86
  | `dimensions` | 768, 512, 256 or 128 | 768 |
87
  | `input_type` | a sentence-transformers prompt name, for example `query`, `document`, `CodeRetrieval` | no prompt |
88
  | `encoding_format` | `float`, or `base64` (little-endian float32) | `float` |
89
- | `model` | optional; `google/embeddinggemma-2` or `tt-hous/embeddinggemma-2-p150` | |
90
 
91
  A media item is an object with optional `text` and at most one media field:
92
 
@@ -160,7 +159,7 @@ Weights under the upstream Apache-2.0 licence (google/embeddinggemma-2). The TTN
160
 
161
  ## Feedback
162
 
163
- Questions or problems with this package: open a discussion at https://huggingface.co/tt-hous/embeddinggemma-2-p150/discussions — that is what reaches its author. A problem with the `tt` tooling itself: `tt report issue` (collects your environment and opens a prefilled issue against tenstorrent/tt-cli). Product feedback: support@tenstorrent.com.
164
 
165
  ## Provenance
166
 
@@ -170,5 +169,5 @@ The exact sources the image was built from — `code/` in this repo is byte-iden
170
  | --- | --- |
171
  | tt-metal | [`3be67656b28e3902f0366ea8bab03c41e7e5bd6c`](https://github.com/tenstorrent/tt-metal/commit/3be67656b28e3902f0366ea8bab03c41e7e5bd6c) |
172
  | `code/` digest | `852ada085de3df45` (sha256, first 16 hex digits) |
173
- | built | 2026-10-21T15:21:44+00:00 by tt-model 0.1.0 |
174
 
 
4
  - p150
5
  - tt-dit-server
6
  - tt-model-cache
 
7
  - tt-model-container
8
  license: apache-2.0
9
  pipeline_tag: feature-extraction
 
45
  ## Quickstart
46
 
47
  ```bash
48
+ tt serve tt-hous/embeddinggemma-2
49
  ```
50
 
51
  `tt serve` downloads the Docker image and the [`google/embeddinggemma-2`](https://huggingface.co/google/embeddinggemma-2) at `914f7f89142e33e77833254d9c9b90c3cef7303b` weights (into your HF cache; they are not in the image), then starts the model's own HTTP server on port 20000 (or the next free port, if that one is busy). The first start compiles kernels for your device, which takes several minutes; the server is ready when it logs `Application startup complete`.
52
 
53
+ Without tt-cli: `tt-model serve tt-hous/embeddinggemma-2`.
54
 
55
  ### Embedding text
56
  ```bash
 
60
 
61
  ### Demo
62
  ```bash
63
+ tt serve tt-hous/embeddinggemma-2 --profile demo
64
  ```
65
  Open `http://<host>:20000/demo/`. A web page where every query is embedded live on the chip, over a bundled library of 163 openly licensed images, videos, sounds, texts in 10 languages and code snippets: media search, finding a moment in a long video, audio search, code search, multilingual search, and a side-by-side Matryoshka comparison.
66
 
 
85
  | `dimensions` | 768, 512, 256 or 128 | 768 |
86
  | `input_type` | a sentence-transformers prompt name, for example `query`, `document`, `CodeRetrieval` | no prompt |
87
  | `encoding_format` | `float`, or `base64` (little-endian float32) | `float` |
88
+ | `model` | optional; `google/embeddinggemma-2` or `embeddinggemma-2` | |
89
 
90
  A media item is an object with optional `text` and at most one media field:
91
 
 
159
 
160
  ## Feedback
161
 
162
+ Questions or problems with this package: open a discussion at https://huggingface.co/tt-hous/embeddinggemma-2/discussions — that is what reaches its author. A problem with the `tt` tooling itself: `tt report issue` (collects your environment and opens a prefilled issue against tenstorrent/tt-cli). Product feedback: support@tenstorrent.com.
163
 
164
  ## Provenance
165
 
 
169
  | --- | --- |
170
  | tt-metal | [`3be67656b28e3902f0366ea8bab03c41e7e5bd6c`](https://github.com/tenstorrent/tt-metal/commit/3be67656b28e3902f0366ea8bab03c41e7e5bd6c) |
171
  | `code/` digest | `852ada085de3df45` (sha256, first 16 hex digits) |
172
+ | built | 2026-10-21T15:23:55+00:00 by tt-model 0.1.0 |
173
 
image/blobs/sha256/0bd59b87edea4b92e96656bdead06a361eac00643fecbca5a1a247c55a460fed ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schemaVersion": 2,
3
+ "mediaType": "application/vnd.oci.image.index.v1+json",
4
+ "manifests": [
5
+ {
6
+ "mediaType": "application/vnd.oci.image.manifest.v1+json",
7
+ "digest": "sha256:b7aba87d343d132afe8e86260c91474534ceabd8928b068eff7e3669a861b3ec",
8
+ "size": 4692,
9
+ "platform": {
10
+ "architecture": "amd64",
11
+ "os": "linux"
12
+ }
13
+ },
14
+ {
15
+ "mediaType": "application/vnd.oci.image.manifest.v1+json",
16
+ "digest": "sha256:398b141b0c8b4ff904b74e545fd23e595a51b12a60a34ffa647311c82b65acf0",
17
+ "size": 564,
18
+ "annotations": {
19
+ "vnd.docker.reference.digest": "sha256:b7aba87d343d132afe8e86260c91474534ceabd8928b068eff7e3669a861b3ec",
20
+ "vnd.docker.reference.type": "attestation-manifest"
21
+ },
22
+ "platform": {
23
+ "architecture": "unknown",
24
+ "os": "unknown"
25
+ }
26
+ }
27
+ ]
28
+ }
image/blobs/sha256/1bfe1becf29e080cde867ea83273cc271f89d4aa5d1e03072d3d4c2023fd1ad5 ADDED
Binary file (6.99 kB). View file
 
image/blobs/sha256/229487199676033e7f4475017c0d8717d719429ce5eeae256e90dd4160c2925f ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:229487199676033e7f4475017c0d8717d719429ce5eeae256e90dd4160c2925f
3
+ size 139115388
image/blobs/sha256/246bf69d316f37f893fdd6ca0e1b6a9dfd2496df809d934747e89f0004dfdfd0 ADDED
Binary file (118 Bytes). View file
 
image/blobs/sha256/3129b0782917119f77fef8dbd067b1bc25dc62520c9b4e0f79d194e33916637a ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3129b0782917119f77fef8dbd067b1bc25dc62520c9b4e0f79d194e33916637a
3
+ size 41630547
image/blobs/sha256/3634bcd964bae60bc52e5812c0ba3d526ff81f0973e0e9aab54aa7bb32a9075d ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3634bcd964bae60bc52e5812c0ba3d526ff81f0973e0e9aab54aa7bb32a9075d
3
+ size 14010550
image/blobs/sha256/398b141b0c8b4ff904b74e545fd23e595a51b12a60a34ffa647311c82b65acf0 ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schemaVersion": 2,
3
+ "mediaType": "application/vnd.oci.image.manifest.v1+json",
4
+ "config": {
5
+ "mediaType": "application/vnd.oci.image.config.v1+json",
6
+ "digest": "sha256:668c29e85693f432d66ce28da65b9880d72f8152431dc87462f2155aa0d29105",
7
+ "size": 167
8
+ },
9
+ "layers": [
10
+ {
11
+ "mediaType": "application/vnd.in-toto+json",
12
+ "digest": "sha256:9dd63fc5eb57989e0662b806e787ab5f6f1be5072b780ef84ae960c9aaf27439",
13
+ "size": 1916,
14
+ "annotations": {
15
+ "in-toto.io/predicate-type": "https://slsa.dev/provenance/v1"
16
+ }
17
+ }
18
+ ]
19
+ }
image/blobs/sha256/3e88ace052c0b2d77a8aec33330ed2387af4a0d4f70ee382ece9faa26739a799 ADDED
Binary file (947 Bytes). View file
 
image/blobs/sha256/44a10ad77f89c27c6dcf8a4489fe0c2cd4bc209d263ccb9c1f3350c5683cf635 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44a10ad77f89c27c6dcf8a4489fe0c2cd4bc209d263ccb9c1f3350c5683cf635
3
+ size 408261884
image/blobs/sha256/468f5474fec5781afd1d4819ed518b57789ecddf1a33c4b26e4ae24b943beb2b ADDED
@@ -0,0 +1 @@
 
 
1
+ {"architecture":"amd64","config":{"User":"tt","Env":["PATH=/opt/tt-venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin","VENV=/opt/tt-venv","VIRTUAL_ENV=/opt/tt-venv","TT_METAL_RUNTIME_ROOT=/opt/tt-metal","TT_METAL_HOME=/opt/tt-metal","PYTHONPATH=/opt/tt-metal","LD_LIBRARY_PATH=/opt/tt-metal/build/lib:/opt/openmpi-v5.0.7-ulfm/lib","EXTRA_MODELS_DIR=","TT_VLLM_BUILTIN_MODELS=","TT_MODEL_KIND=tt-dit-server","HF_HOME=/hf","TT_METAL_CACHE=/cache","HOME=/home/tt","USER=tt","LOGNAME=tt"],"Entrypoint":["/usr/local/bin/entrypoint.sh"],"Cmd":["/usr/local/bin/serve-default.sh"],"WorkingDir":"/home/tt/work","Labels":{"org.opencontainers.image.revision":"3be67656b28e3902f0366ea8bab03c41e7e5bd6c","org.opencontainers.image.version":"22.04","org.tenstorrent.tt-model":"embeddinggemma-2-p150","org.tenstorrent.tt-model.arch":"blackhole","org.tenstorrent.tt-model.kind":"tt-dit-server","org.tenstorrent.tt-model.plugin":"","org.tenstorrent.tt-model.profiles":"default,demo","org.tenstorrent.tt-model.repo":"tt-hous/embeddinggemma-2","org.tenstorrent.tt-model.tt-metal":"","org.tenstorrent.tt-model.weights":"google/embeddinggemma-2"},"ArgsEscaped":true},"created":"2026-10-21T15:24:24.452558123Z","history":[{"created":"2026-10-02T10:21:04.163019566Z","created_by":"/bin/sh -c #(nop) ARG RELEASE","empty_layer":true},{"created":"2026-10-02T10:21:04.191260039Z","created_by":"/bin/sh -c #(nop) ARG LAUNCHPAD_BUILD_ARCH","empty_layer":true},{"created":"2026-10-02T10:21:04.236927516Z","created_by":"/bin/sh -c #(nop) LABEL org.opencontainers.image.version=22.04","empty_layer":true},{"created":"2026-10-02T10:21:06.473069851Z","created_by":"/bin/sh -c #(nop) ADD file:b6c410e935243e5623135210c34358091e8af5ff14851cb0f67195733b55d391 in / "},{"created":"2026-10-02T10:21:06.943369844Z","created_by":"/bin/sh -c #(nop) CMD [\"/bin/bash\"]","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG OMPI_DIR=/opt/openmpi-v5.0.7-ulfm","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG EXTRA_MODELS_DIR=","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG TT_MODEL_KIND","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_NAME","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_REPO","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_WEIGHTS","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_ARCH","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_PROFILES","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_TT_METAL_SHA","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_TT_METAL_DESCRIBE","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"ARG MODEL_PLUGIN_SHA","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:01.229012302Z","created_by":"RUN |11 OMPI_DIR=/opt/openmpi-v5.0.7-ulfm EXTRA_MODELS_DIR= TT_MODEL_KIND=tt-dit-server MODEL_NAME=embeddinggemma-2-p150 MODEL_REPO=tt-hous/embeddinggemma-2 MODEL_WEIGHTS=google/embeddinggemma-2 MODEL_ARCH=blackhole MODEL_PROFILES=default,demo MODEL_TT_METAL_SHA=3be67656b28e3902f0366ea8bab03c41e7e5bd6c MODEL_TT_METAL_DESCRIBE= MODEL_PLUGIN_SHA= /bin/sh -c apt-get update \u0026\u0026 apt-get install -y --no-install-recommends libhwloc15 libnuma1 libatomic1 libudev1 libcap2 zlib1g libmpc3 libmpfr6 libgmp10 libzstd1 libevent-core-2.1-7 libevent-pthreads-2.1-7 libgl1 libsndfile1 ca-certificates \u0026\u0026 apt-get clean \u0026\u0026 rm -rf /var/lib/apt/lists/* # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:01.361134766Z","created_by":"RUN |11 OMPI_DIR=/opt/openmpi-v5.0.7-ulfm EXTRA_MODELS_DIR= TT_MODEL_KIND=tt-dit-server MODEL_NAME=embeddinggemma-2-p150 MODEL_REPO=tt-hous/embeddinggemma-2 MODEL_WEIGHTS=google/embeddinggemma-2 MODEL_ARCH=blackhole MODEL_PROFILES=default,demo MODEL_TT_METAL_SHA=3be67656b28e3902f0366ea8bab03c41e7e5bd6c MODEL_TT_METAL_DESCRIBE= MODEL_PLUGIN_SHA= /bin/sh -c existing=\"$(getent passwd 1000 | cut -d: -f1)\" \u0026\u0026 if [ -n \"$existing\" ]; then userdel -r \"$existing\" 2\u003e/dev/null || userdel \"$existing\"; fi \u0026\u0026 useradd --uid 1000 --create-home --home-dir /home/tt --shell /bin/bash tt \u0026\u0026 mkdir -p /home/tt/work/logs /cache /opt/tt-metal \u0026\u0026 chown -R tt:tt /home/tt /cache /opt/tt-metal \u0026\u0026 chmod 1777 /home/tt /home/tt/work /home/tt/work/logs /cache # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:01.403739898Z","created_by":"COPY /opt/openmpi-v5.0.7-ulfm /opt/openmpi-v5.0.7-ulfm # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:01.620618108Z","created_by":"COPY /opt/tenstorrent /opt/tenstorrent # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:01.931574695Z","created_by":"COPY /usr/local/share/uv /usr/local/share/uv # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:03.212612979Z","created_by":"COPY /opt/tt-venv /opt/tt-venv # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:03.288857254Z","created_by":"COPY /opt/vllm /opt/vllm # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:03.500309503Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/runtime /opt/tt-metal/runtime # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:03.570313769Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/build_Release /opt/tt-metal/build_Release # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:03.657905258Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/build /opt/tt-metal/build # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:03.938326708Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/tt_metal /opt/tt-metal/tt_metal # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.177829646Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/ttnn /opt/tt-metal/ttnn # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.224143696Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/tools /opt/tt-metal/tools # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.238469333Z","created_by":"COPY --chown=tt:tt /opt/tt-metal/setup.py /opt/tt-metal/pyproject.toml /opt/tt-metal/ # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.272044724Z","created_by":"COPY --chown=tt:tt code/ /opt/tt-metal/ # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.504391115Z","created_by":"RUN |11 OMPI_DIR=/opt/openmpi-v5.0.7-ulfm EXTRA_MODELS_DIR= TT_MODEL_KIND=tt-dit-server MODEL_NAME=embeddinggemma-2-p150 MODEL_REPO=tt-hous/embeddinggemma-2 MODEL_WEIGHTS=google/embeddinggemma-2 MODEL_ARCH=blackhole MODEL_PROFILES=default,demo MODEL_TT_METAL_SHA=3be67656b28e3902f0366ea8bab03c41e7e5bd6c MODEL_TT_METAL_DESCRIBE= MODEL_PLUGIN_SHA= /bin/sh -c find /opt/tt-metal \\( -type f -o -type d \\) \\( ! -perm -o+r -o \\( -type d ! -perm -o+x \\) \\) -exec chmod a+rX {} + # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.518920862Z","created_by":"COPY --chmod=0755 entrypoint.sh /usr/local/bin/entrypoint.sh # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"COPY --chmod=0755 serve-default.sh /usr/local/bin/serve-default.sh # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV VENV=/opt/tt-venv","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV VIRTUAL_ENV=/opt/tt-venv","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV PATH=/opt/tt-venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV TT_METAL_RUNTIME_ROOT=/opt/tt-metal","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV TT_METAL_HOME=/opt/tt-metal","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV PYTHONPATH=/opt/tt-metal","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV LD_LIBRARY_PATH=/opt/tt-metal/build/lib:/opt/openmpi-v5.0.7-ulfm/lib","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV EXTRA_MODELS_DIR=","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ARG TT_VLLM_BUILTIN_MODELS=","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV TT_VLLM_BUILTIN_MODELS=","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV TT_MODEL_KIND=tt-dit-server","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV HF_HOME=/hf","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV TT_METAL_CACHE=/cache","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV HOME=/home/tt","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV USER=tt","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"ENV LOGNAME=tt","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.528597568Z","created_by":"USER tt","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:04.539213726Z","created_by":"WORKDIR /home/tt/work","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:04.551334516Z","created_by":"COPY --chmod=0755 verify.sh /ctx/verify.sh # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:24.337890376Z","created_by":"RUN |12 OMPI_DIR=/opt/openmpi-v5.0.7-ulfm EXTRA_MODELS_DIR= TT_MODEL_KIND=tt-dit-server MODEL_NAME=embeddinggemma-2-p150 MODEL_REPO=tt-hous/embeddinggemma-2 MODEL_WEIGHTS=google/embeddinggemma-2 MODEL_ARCH=blackhole MODEL_PROFILES=default,demo MODEL_TT_METAL_SHA=3be67656b28e3902f0366ea8bab03c41e7e5bd6c MODEL_TT_METAL_DESCRIBE= MODEL_PLUGIN_SHA= TT_VLLM_BUILTIN_MODELS= /bin/sh -c bash /ctx/verify.sh # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:24.337890376Z","created_by":"USER root","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:24.452558123Z","created_by":"RUN |12 OMPI_DIR=/opt/openmpi-v5.0.7-ulfm EXTRA_MODELS_DIR= TT_MODEL_KIND=tt-dit-server MODEL_NAME=embeddinggemma-2-p150 MODEL_REPO=tt-hous/embeddinggemma-2 MODEL_WEIGHTS=google/embeddinggemma-2 MODEL_ARCH=blackhole MODEL_PROFILES=default,demo MODEL_TT_METAL_SHA=3be67656b28e3902f0366ea8bab03c41e7e5bd6c MODEL_TT_METAL_DESCRIBE= MODEL_PLUGIN_SHA= TT_VLLM_BUILTIN_MODELS= /bin/sh -c chmod -R a+rwX /home/tt # buildkit","comment":"buildkit.dockerfile.v0"},{"created":"2026-10-21T15:24:24.452558123Z","created_by":"USER tt","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:24.452558123Z","created_by":"LABEL org.tenstorrent.tt-model=embeddinggemma-2-p150 org.tenstorrent.tt-model.repo=tt-hous/embeddinggemma-2 org.tenstorrent.tt-model.weights=google/embeddinggemma-2 org.tenstorrent.tt-model.arch=blackhole org.tenstorrent.tt-model.kind=tt-dit-server org.tenstorrent.tt-model.profiles=default,demo org.opencontainers.image.revision=3be67656b28e3902f0366ea8bab03c41e7e5bd6c org.tenstorrent.tt-model.tt-metal= org.tenstorrent.tt-model.plugin=","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:24.452558123Z","created_by":"ENTRYPOINT [\"/usr/local/bin/entrypoint.sh\"]","comment":"buildkit.dockerfile.v0","empty_layer":true},{"created":"2026-10-21T15:24:24.452558123Z","created_by":"CMD [\"/usr/local/bin/serve-default.sh\"]","comment":"buildkit.dockerfile.v0","empty_layer":true}],"os":"linux","rootfs":{"type":"layers","diff_ids":["sha256:129f77d88403f1d66805466da50673bab10cf6052bdf1fe422b0341b51a55da2","sha256:df238dccdeb97975e444e0737e4a5cf74df55c2d27b9bea45f9e2066e344e156","sha256:4c8193a2d19eaa1c22f684ad9778f9faf7dc7a89dbaf2fd7c2d2a40a3151e972","sha256:89b7907439b149369d689bc869d850640fc0412c1dff1406d7ef732e057acb8f","sha256:2286eed2f7146ca665ed3a5bd9aa289dca83ee58ce37de3c29ec909fcc922ede","sha256:baa0fb2e9ac74ae208ebe38aae200bc5cd3ecadd7fa8c3d0098b6934cc5a0255","sha256:3cf323a065e2f9e0797b529d4f81aee201a2733f336977bdbce3616c9f074b31","sha256:b442151eafb14a12f3fbe68ef86fdf4f78e730b6d308ca2b2d4e94793c925a96","sha256:e00744c1ebf7029da9c61eb05ad48626fb441a89522732509f65cd6d895b5af7","sha256:0938002959e16773048eb211c5162cf3dbb65cc2de4c179dc1dc3bd4041e1477","sha256:c516140dafe60d6ce66edcc1362710b22918f54a79baa4e7c8e9b0a1ad7ce8de","sha256:5306a299d814b1a208b367c5f5c95f151c19992ef6acefb23a81703b07cc384e","sha256:d36244291314abb998d52e9a530ceb24ecd6dbceef695ba2a59b361cd8c2e15e","sha256:5755986846fe857e4190703d6d94a68611741de70b15a8e3c09cfc4807927b29","sha256:170a2b563a1ada483e41aa5d6f3019430b87a8167d71bd6d2847c46d04cacea8","sha256:dd17be66c790880937fa129404d44f4b10d345b928447e3a4c1731b2219af15d","sha256:4bfd28e3942a3eb3fa0e7dde8ceff6af2dd83a73136378ff787122866cf6b352","sha256:6221d31e2787443210a766f73a5868aa1d55fb12ba680308f6a0b3b26c2158af","sha256:621725bb4076c2f2bc26b5401aa6e79ccda50241bd40926f11d975562a80b2ff","sha256:5f70bf18a086007016e948b04aed3b82103a36bea41755b6cddfaf10ace3c6ef","sha256:323a380bc02b1b73f610eb20fbb750d5cad16ee1dec59431ed46f9e2f1860645","sha256:a95b3bc18402d93cd1a3e397acc4e1b113bdc7022c798d57f15b8238a5b87946","sha256:1f7456e77868213ae16c821a9d2e93e1654503a3a36aba3d02afc0e64ec26bd8"]}}
image/blobs/sha256/4b30923806c45caff71513ec3e81797253f5bc240d27874612b4bbccae95efad ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b30923806c45caff71513ec3e81797253f5bc240d27874612b4bbccae95efad
3
+ size 41970940
image/blobs/sha256/4bbd042d5a48f78d42374a9ff7099c61ef4155fce27d055a3e5d572ea2887230 ADDED
Binary file (2.32 kB). View file
 
image/blobs/sha256/510954577bc5f9942a88171596d66e859741843fbe9eaac1f58e6268baf88c5e ADDED
Binary file (4.4 kB). View file
 
image/blobs/sha256/5a009e1245fa4110a9b1e473cd2feba0a201d8d85dbec0941269e4592d27c506 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a009e1245fa4110a9b1e473cd2feba0a201d8d85dbec0941269e4592d27c506
3
+ size 41970721
image/blobs/sha256/5d85ec83334f0a3af355a319fbf5bbf743180ace7d77fe9081a84a78cc2f3f3d ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d85ec83334f0a3af355a319fbf5bbf743180ace7d77fe9081a84a78cc2f3f3d
3
+ size 67583604
image/blobs/sha256/65de262429bc470f1e6c7de03c08d57a9eb77a30ba82d226fecce3aafbeb306c ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65de262429bc470f1e6c7de03c08d57a9eb77a30ba82d226fecce3aafbeb306c
3
+ size 69556776
image/blobs/sha256/668c29e85693f432d66ce28da65b9880d72f8152431dc87462f2155aa0d29105 ADDED
@@ -0,0 +1 @@
 
 
1
+ {"architecture":"unknown","os":"unknown","config":{},"rootfs":{"type":"layers","diff_ids":["sha256:9dd63fc5eb57989e0662b806e787ab5f6f1be5072b780ef84ae960c9aaf27439"]}}
image/blobs/sha256/9dd63fc5eb57989e0662b806e787ab5f6f1be5072b780ef84ae960c9aaf27439 ADDED
@@ -0,0 +1 @@
 
 
1
+ {"_type":"https://in-toto.io/Statement/v0.1","predicateType":"https://slsa.dev/provenance/v1","subject":[{"name":"pkg:docker/tt-model/embeddinggemma-2-p150@build-3be67656b?platform=linux%2Famd64","digest":{"sha256":"b7aba87d343d132afe8e86260c91474534ceabd8928b068eff7e3669a861b3ec"}}],"predicate":{"buildDefinition":{"buildType":"https://github.com/moby/buildkit/blob/master/docs/attestations/slsa-definitions.md","resolvedDependencies":[{"uri":"pkg:docker/docker/dockerfile@1.7-labs","digest":{"sha256":"b99fecfe00268a8b556fad7d9c37ee25d716ae08a5d7320e6d51c4dd83246894"}},{"uri":"pkg:docker/ubuntu@22.04?platform=linux%2Famd64","digest":{"sha256":"0a9887931b0e38af2e3487ed39e29436b0535c4f3f52e57745baaf79d1898956"}},{"uri":"pkg:docker/ghcr.io/tenstorrent/tt-metal/tt-metalium/ubuntu-22.04-dev-amd64@latest?platform=linux%2Famd64","digest":{"sha256":"16bcf2a1e76051d857a1bee935b589fa4a9b73ee1a9e87eee4ce063de3cd2448"}}],"externalParameters":{"configSource":{"path":"Dockerfile"},"request":{"frontend":"gateway.v0","args":{"cmdline":"docker/dockerfile:1.7-labs","context:metalsrc":"local:metalsrc","frontend.caps":"moby.buildkit.frontend.contexts+forward","sharedkey:localdir:metalsrc":"empty-metalsrc:98849442d91b4194","source":"docker/dockerfile:1.7-labs"},"locals":[{"name":"context"},{"name":"dockerfile"},{"name":"metalsrc"}],"root":{"configSource":{"path":"Dockerfile"},"request":{"args":{"context:metalsrc":"local:metalsrc","frontend.caps":"moby.buildkit.frontend.contexts+forward","sharedkey:localdir:metalsrc":"empty-metalsrc:98849442d91b4194"}}},"compatibilityVersion":20}},"internalParameters":{"builderPlatform":"linux/amd64"}},"runDetails":{"builder":{"id":""},"metadata":{"invocationId":"oa4l1umqjvz9d7e8qfqcir5ga","startedOn":"2026-10-21T15:23:55.901184995Z","finishedOn":"2026-10-21T15:24:48.289691973Z","buildkit_metadata":{},"buildkit_completeness":{"request":false,"resolvedDependencies":false}}}}}
image/blobs/sha256/b349c5f82ec7e1c9285f4c41a294f1aa9efebbd757791dc3712f4b869d2b4cd5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b349c5f82ec7e1c9285f4c41a294f1aa9efebbd757791dc3712f4b869d2b4cd5
3
+ size 41630392
image/blobs/sha256/b7aba87d343d132afe8e86260c91474534ceabd8928b068eff7e3669a861b3ec ADDED
@@ -0,0 +1,126 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schemaVersion": 2,
3
+ "mediaType": "application/vnd.oci.image.manifest.v1+json",
4
+ "config": {
5
+ "mediaType": "application/vnd.oci.image.config.v1+json",
6
+ "digest": "sha256:468f5474fec5781afd1d4819ed518b57789ecddf1a33c4b26e4ae24b943beb2b",
7
+ "size": 14712
8
+ },
9
+ "layers": [
10
+ {
11
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
12
+ "digest": "sha256:cf1a5487baaf7d71eb78fd0888933c6cdecb40e59edb896511aa0ef4a896031e",
13
+ "size": 29752476
14
+ },
15
+ {
16
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
17
+ "digest": "sha256:65de262429bc470f1e6c7de03c08d57a9eb77a30ba82d226fecce3aafbeb306c",
18
+ "size": 69556776
19
+ },
20
+ {
21
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
22
+ "digest": "sha256:510954577bc5f9942a88171596d66e859741843fbe9eaac1f58e6268baf88c5e",
23
+ "size": 4396
24
+ },
25
+ {
26
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
27
+ "digest": "sha256:e10f09843c9b4d73e9fd19f406e6bd9634c867ccdc92552898cfc760fc56ef6f",
28
+ "size": 9565639
29
+ },
30
+ {
31
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
32
+ "digest": "sha256:f38ef5e3e335490fe8905045047f1bd8276adc0add86089deceb7151d573dd42",
33
+ "size": 138999285
34
+ },
35
+ {
36
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
37
+ "digest": "sha256:be74ff0f8fffc8c250798344d8e2fa17f6a6a69c79586273b64b1fcf4b55c713",
38
+ "size": 66025129
39
+ },
40
+ {
41
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
42
+ "digest": "sha256:44a10ad77f89c27c6dcf8a4489fe0c2cd4bc209d263ccb9c1f3350c5683cf635",
43
+ "size": 408261884
44
+ },
45
+ {
46
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
47
+ "digest": "sha256:246bf69d316f37f893fdd6ca0e1b6a9dfd2496df809d934747e89f0004dfdfd0",
48
+ "size": 118
49
+ },
50
+ {
51
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
52
+ "digest": "sha256:229487199676033e7f4475017c0d8717d719429ce5eeae256e90dd4160c2925f",
53
+ "size": 139115388
54
+ },
55
+ {
56
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
57
+ "digest": "sha256:5a009e1245fa4110a9b1e473cd2feba0a201d8d85dbec0941269e4592d27c506",
58
+ "size": 41970721
59
+ },
60
+ {
61
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
62
+ "digest": "sha256:4b30923806c45caff71513ec3e81797253f5bc240d27874612b4bbccae95efad",
63
+ "size": 41970940
64
+ },
65
+ {
66
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
67
+ "digest": "sha256:5d85ec83334f0a3af355a319fbf5bbf743180ace7d77fe9081a84a78cc2f3f3d",
68
+ "size": 67583604
69
+ },
70
+ {
71
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
72
+ "digest": "sha256:3634bcd964bae60bc52e5812c0ba3d526ff81f0973e0e9aab54aa7bb32a9075d",
73
+ "size": 14010550
74
+ },
75
+ {
76
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
77
+ "digest": "sha256:d7161b2ea02a42b9f3b038066bb83ede992815f0461a5e35644ca91594aa6ac6",
78
+ "size": 1967440
79
+ },
80
+ {
81
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
82
+ "digest": "sha256:1bfe1becf29e080cde867ea83273cc271f89d4aa5d1e03072d3d4c2023fd1ad5",
83
+ "size": 6988
84
+ },
85
+ {
86
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
87
+ "digest": "sha256:b349c5f82ec7e1c9285f4c41a294f1aa9efebbd757791dc3712f4b869d2b4cd5",
88
+ "size": 41630392
89
+ },
90
+ {
91
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
92
+ "digest": "sha256:3129b0782917119f77fef8dbd067b1bc25dc62520c9b4e0f79d194e33916637a",
93
+ "size": 41630547
94
+ },
95
+ {
96
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
97
+ "digest": "sha256:df20c74836f348e00350c8f20fc9ec5948ed47ebdaaa0321afa85de40d341fab",
98
+ "size": 1080
99
+ },
100
+ {
101
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
102
+ "digest": "sha256:d853bc77781c2b20971eb04b2a3bd79e993f899c6ac962b5211ba7bf81064a22",
103
+ "size": 1079
104
+ },
105
+ {
106
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
107
+ "digest": "sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1",
108
+ "size": 32
109
+ },
110
+ {
111
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
112
+ "digest": "sha256:3e88ace052c0b2d77a8aec33330ed2387af4a0d4f70ee382ece9faa26739a799",
113
+ "size": 947
114
+ },
115
+ {
116
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
117
+ "digest": "sha256:b891528e3c01140e7ab06a4836d79cff05712ae577e3b798c81b5de12095d35b",
118
+ "size": 267857
119
+ },
120
+ {
121
+ "mediaType": "application/vnd.oci.image.layer.v1.tar+gzip",
122
+ "digest": "sha256:4bbd042d5a48f78d42374a9ff7099c61ef4155fce27d055a3e5d572ea2887230",
123
+ "size": 2323
124
+ }
125
+ ]
126
+ }
image/blobs/sha256/b891528e3c01140e7ab06a4836d79cff05712ae577e3b798c81b5de12095d35b ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b891528e3c01140e7ab06a4836d79cff05712ae577e3b798c81b5de12095d35b
3
+ size 267857
image/blobs/sha256/be74ff0f8fffc8c250798344d8e2fa17f6a6a69c79586273b64b1fcf4b55c713 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be74ff0f8fffc8c250798344d8e2fa17f6a6a69c79586273b64b1fcf4b55c713
3
+ size 66025129
image/blobs/sha256/d7161b2ea02a42b9f3b038066bb83ede992815f0461a5e35644ca91594aa6ac6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7161b2ea02a42b9f3b038066bb83ede992815f0461a5e35644ca91594aa6ac6
3
+ size 1967440
image/blobs/sha256/d853bc77781c2b20971eb04b2a3bd79e993f899c6ac962b5211ba7bf81064a22 ADDED
Binary file (1.08 kB). View file
 
image/blobs/sha256/df20c74836f348e00350c8f20fc9ec5948ed47ebdaaa0321afa85de40d341fab ADDED
Binary file (1.08 kB). View file
 
image/blobs/sha256/e10f09843c9b4d73e9fd19f406e6bd9634c867ccdc92552898cfc760fc56ef6f ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e10f09843c9b4d73e9fd19f406e6bd9634c867ccdc92552898cfc760fc56ef6f
3
+ size 9565639
image/blobs/sha256/f38ef5e3e335490fe8905045047f1bd8276adc0add86089deceb7151d573dd42 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f38ef5e3e335490fe8905045047f1bd8276adc0add86089deceb7151d573dd42
3
+ size 138999285
image/index.json CHANGED
@@ -1 +1 @@
1
- {"schemaVersion":2,"mediaType":"application/vnd.oci.image.index.v1+json","manifests":[{"mediaType":"application/vnd.oci.image.index.v1+json","digest":"sha256:bee93e165d00114d3ba68d68a650f6898a743bda5c6c00bff50c0540b20f25f4","size":856,"annotations":{"io.containerd.image.name":"docker.io/tt-model/embeddinggemma-2-p150:bee93e165d00","org.opencontainers.image.ref.name":"bee93e165d00"}}]}
 
1
+ {"schemaVersion":2,"mediaType":"application/vnd.oci.image.index.v1+json","manifests":[{"mediaType":"application/vnd.oci.image.index.v1+json","digest":"sha256:0bd59b87edea4b92e96656bdead06a361eac00643fecbca5a1a247c55a460fed","size":856,"annotations":{"io.containerd.image.name":"docker.io/tt-model/embeddinggemma-2-p150:0bd59b87edea","org.opencontainers.image.ref.name":"0bd59b87edea"}}]}
image/manifest.json CHANGED
@@ -1 +1 @@
1
- [{"Config":"blobs/sha256/9a9e80da4de6221fd880dc96810376e07d22c2984bb2628ced2e8d0b9c8cd7de","RepoTags":["tt-model/embeddinggemma-2-p150:bee93e165d00"],"Layers":["blobs/sha256/cf1a5487baaf7d71eb78fd0888933c6cdecb40e59edb896511aa0ef4a896031e","blobs/sha256/11d2eac53ff4ff7d3d67057613e4f204fe7da40b8ba001e09a7b785a69827da3","blobs/sha256/35e14da4db1e99991e4ccc3a0d01584086548bae7120c5c368ae5c131f28121a","blobs/sha256/432d2195132d36370aca685dae5eb2d62628dd8f0214ba236618f84198620aa2","blobs/sha256/46713da844e7a0c5aa5f5c79e7270637bf751186078ec5bc260dd7b06006242a","blobs/sha256/451dddf7926009d9f2d3e471e6b27af1af62b2e86a6c061738bdab2b9a4c1a56","blobs/sha256/b5ec001ffd467834132e55397d625249decf5f5f63079307e34ae3acc00e8436","blobs/sha256/b749e800205c3ba3769ba8a199bd85d353f376391ad5f68f16da895d126b82d0","blobs/sha256/a476bdec81d16f5259e66edb2670af7383593c49a98b6cec2427117d756e11c0","blobs/sha256/aba408305578d428cf39127b2a6ce91037c83b6c33f08497174df12843f72b94","blobs/sha256/b65b506905faa6bdd4125549d97257e64c8a5715c11c9780b6a9b7678fd05f07","blobs/sha256/56e26571152696889aa2d2dca1568ff6c70030b12c08b5f196a577ea1cf1afea","blobs/sha256/59abbc84c8a3402b41571973de30f9d6f5e8aeac1919e155602e3570e12995fd","blobs/sha256/fa71074b04af7e9df6466c49c50e461c6f6e1d986900d94092fbb3994a598a53","blobs/sha256/4006fa78a01aa3b5f6051dca103138a25743d828059be8ee4dd4ecca184e78ff","blobs/sha256/75541aedee66d279bb9db97e12a0a82fef223f9104d3e3dcb259a8104e489d8a","blobs/sha256/9a7118cb3c11a1cd38db3e354ad23fefe52a05396021c832422810cbc832650b","blobs/sha256/3751585d3eb5789fa5fec521a6176185ab2299273f215f825f6b0d54b873248e","blobs/sha256/f4ebe6bafc19cd6b0f41fc09321cd0b2676884ebd3b06d95deac3c2f0786fdc2","blobs/sha256/4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1","blobs/sha256/2b553ee3860d69cd0ca80aa40607f7b6f3bf5110b0aa428deeee6a57110f8ab0","blobs/sha256/c71cc42b78a54d49663258ad776fe0ceb6402a360b1b699e6d709464dfec3597","blobs/sha256/0f607516b935aed5184f195141354b0c6cd0bee121742159a513be94baaa48eb"]}]
 
1
+ [{"Config":"blobs/sha256/468f5474fec5781afd1d4819ed518b57789ecddf1a33c4b26e4ae24b943beb2b","RepoTags":["tt-model/embeddinggemma-2-p150:0bd59b87edea"],"Layers":["blobs/sha256/cf1a5487baaf7d71eb78fd0888933c6cdecb40e59edb896511aa0ef4a896031e","blobs/sha256/65de262429bc470f1e6c7de03c08d57a9eb77a30ba82d226fecce3aafbeb306c","blobs/sha256/510954577bc5f9942a88171596d66e859741843fbe9eaac1f58e6268baf88c5e","blobs/sha256/e10f09843c9b4d73e9fd19f406e6bd9634c867ccdc92552898cfc760fc56ef6f","blobs/sha256/f38ef5e3e335490fe8905045047f1bd8276adc0add86089deceb7151d573dd42","blobs/sha256/be74ff0f8fffc8c250798344d8e2fa17f6a6a69c79586273b64b1fcf4b55c713","blobs/sha256/44a10ad77f89c27c6dcf8a4489fe0c2cd4bc209d263ccb9c1f3350c5683cf635","blobs/sha256/246bf69d316f37f893fdd6ca0e1b6a9dfd2496df809d934747e89f0004dfdfd0","blobs/sha256/229487199676033e7f4475017c0d8717d719429ce5eeae256e90dd4160c2925f","blobs/sha256/5a009e1245fa4110a9b1e473cd2feba0a201d8d85dbec0941269e4592d27c506","blobs/sha256/4b30923806c45caff71513ec3e81797253f5bc240d27874612b4bbccae95efad","blobs/sha256/5d85ec83334f0a3af355a319fbf5bbf743180ace7d77fe9081a84a78cc2f3f3d","blobs/sha256/3634bcd964bae60bc52e5812c0ba3d526ff81f0973e0e9aab54aa7bb32a9075d","blobs/sha256/d7161b2ea02a42b9f3b038066bb83ede992815f0461a5e35644ca91594aa6ac6","blobs/sha256/1bfe1becf29e080cde867ea83273cc271f89d4aa5d1e03072d3d4c2023fd1ad5","blobs/sha256/b349c5f82ec7e1c9285f4c41a294f1aa9efebbd757791dc3712f4b869d2b4cd5","blobs/sha256/3129b0782917119f77fef8dbd067b1bc25dc62520c9b4e0f79d194e33916637a","blobs/sha256/df20c74836f348e00350c8f20fc9ec5948ed47ebdaaa0321afa85de40d341fab","blobs/sha256/d853bc77781c2b20971eb04b2a3bd79e993f899c6ac962b5211ba7bf81064a22","blobs/sha256/4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1","blobs/sha256/3e88ace052c0b2d77a8aec33330ed2387af4a0d4f70ee382ece9faa26739a799","blobs/sha256/b891528e3c01140e7ab06a4836d79cff05712ae577e3b798c81b5de12095d35b","blobs/sha256/4bbd042d5a48f78d42374a9ff7099c61ef4155fce27d055a3e5d572ea2887230"]}]
tt_kernel_manifest.json CHANGED
@@ -6,7 +6,7 @@
6
  "device_count": 1,
7
  "producer": {
8
  "tt_kernel_version": "0.1.0",
9
- "created_at": "2026-10-21T15:21:48.898834+00:00",
10
  "hostname": "qb2-120-p06t08"
11
  },
12
  "weights": {
@@ -28,8 +28,8 @@
28
  "image": {
29
  "registry": "hf",
30
  "repository": "embeddinggemma-2-p150",
31
- "tag": "tt-model/embeddinggemma-2-p150:bee93e165d00",
32
- "digest": "sha256:bee93e165d00114d3ba68d68a650f6898a743bda5c6c00bff50c0540b20f25f4"
33
  },
34
  "kind": "tt-dit-server",
35
  "runtime": {
@@ -102,10 +102,10 @@
102
  "import gradio, json, models.autoports.google_embeddinggemma_2.demo.ui as u, models.autoports.google_embeddinggemma_2.demo.index as i; assert gradio.__version__ == '6.30.0', gradio.__version__; m = json.loads((i.LIBRARY_DIR / 'manifest.json').read_text()); missing = [x['file'] for x in m['items'] if x.get('file') and not (i.LIBRARY_DIR / x['file']).is_file()]; assert not missing, missing[:5]"
103
  ],
104
  "built": {
105
- "image": "tt-model/embeddinggemma-2-p150:bee93e165d00",
106
- "repo": "tt-hous/embeddinggemma-2-p150",
107
  "tt_model_version": "0.1.0",
108
- "created_at": "2026-10-21T15:21:44+00:00",
109
  "tt_metal": {
110
  "sha": "3be67656b28e3902f0366ea8bab03c41e7e5bd6c",
111
  "describe": null,
@@ -117,18 +117,18 @@
117
  "pushed": null
118
  },
119
  "code_sha256": "852ada085de3df458409d2183b853e843234bd60acff1efb98eb3c32c62dd77d",
120
- "image_digest": "sha256:bee93e165d00114d3ba68d68a650f6898a743bda5c6c00bff50c0540b20f25f4"
121
  },
122
  "card": {
123
  "description": "EmbeddingGemma 2, a multimodal embedding model. Text, code, images, video frames and 16 kHz audio each map to one L2-normalized 768-dimensional vector in a shared space, so a text query can find an image, a sound or a video. Matryoshka dimensions 512, 256 and 128 are the first values of the 768-dimensional vector, re-normalized on the chip.\n\nThe text tower is a bidirectional 24-layer Gemma encoder (hidden size 512, sliding and full attention, 8192-token context). Gemma 4 vision (16 layers) and audio (12 layers) towers feed soft tokens into it. Weights stay on the chip and every request shape runs as a captured TTNN trace. Two custom kernels (sliding-window SDPA and a GEGLU matmul) run through stock `ttnn.generic_op`; tt-metal itself is unmodified.\n",
124
  "attribution": "Derived from google/embeddinggemma-2 by Google DeepMind, weights unmodified. The TTNN port is models/autoports/google_embeddinggemma_2, on stock upstream tt-metal.\n",
125
  "prerequisites": "- 1 GB hugepages mounted at `/dev/hugepages-1G`\n- One chip of a p300 also works: the package sets the p150 mesh graph descriptor.\n",
126
- "quickstart": "### Embedding text\n```bash\ncurl -s localhost:20000/v1/embeddings -H 'content-type: application/json' \\\n -d '{\"input\": [\"What is the capital of France?\"], \"input_type\": \"query\", \"dimensions\": 256}'\n```\n\n### Demo\n```bash\ntt serve tt-hous/embeddinggemma-2-p150 --profile demo\n```\nOpen `http://<host>:20000/demo/`. A web page where every query is embedded live on the chip, over a bundled library of 163 openly licensed images, videos, sounds, texts in 10 languages and code snippets: media search, finding a moment in a long video, audio search, code search, multilingual search, and a side-by-side Matryoshka comparison.\n",
127
  "architecture": null,
128
  "status": null,
129
  "intended_use": "Embeddings for retrieval, semantic search, clustering and classification over text, code, images, video and audio, served from one p150. Not for generating text.\n",
130
  "out_of_scope_use": null,
131
- "usage": "`POST /v1/embeddings` follows the OpenAI embeddings request and response shapes, with media extensions.\n\n| field | meaning | default |\n|---|---|---|\n| `input` | a string, or a list of 1 to 2048 strings and media items | required |\n| `dimensions` | 768, 512, 256 or 128 | 768 |\n| `input_type` | a sentence-transformers prompt name, for example `query`, `document`, `CodeRetrieval` | no prompt |\n| `encoding_format` | `float`, or `base64` (little-endian float32) | `float` |\n| `model` | optional; `google/embeddinggemma-2` or `tt-hous/embeddinggemma-2-p150` | |\n\nA media item is an object with optional `text` and at most one media field:\n\n| field | value |\n|---|---|\n| `image` | base64 image bytes (any PIL format) or a `data:image/...;base64,` URL |\n| `audio` | base64 16 kHz WAV; clips over 30 s are cut to 30 s |\n| `video` | `{\"frames\": [<base64 image>, ...], \"fps\": 1}`; at most 32 frames are used |\n\n```bash\nIMG=$(base64 -w0 cats.jpg)\ncurl -s localhost:20000/v1/embeddings -H 'content-type: application/json' \\\n -d \"{\\\"input\\\": [{\\\"image\\\": \\\"$IMG\\\"}, {\\\"text\\\": \\\"two cats on a couch\\\"}]}\"\n```\n\n`GET /health` answers 200 once the traces are captured, with the precision policy and live counters, and 503 before that. `GET /v1/models` lists the model, its dimensions and modalities. Errors use the OpenAI error shape; an input over 8192 tokens gets 400 `context_length_exceeded`. Media URLs are refused: send media as base64.\n",
132
  "performance": "| benchmark (768d) | metric | this package | HF fp32 | card |\n| --- | --- | ---: | ---: | ---: |\n| MTEB multilingual v2 | Mean(Task) | 59.55 | 59.51 | 61.36 |\n| MTEB eng v2 | Mean(Task) | 70.25 | 70.35 | 68.46 |\n| MTEB code v1 | Mean(Task), NDCG@10 | 56.99 | 57.04 | 78.68 |\n| MIEB lite | Mean(TaskType) | 60.93 | 60.50 | 64.64 |\n| MMEB v2 Image | Mean(Task), Hit@1 | 38.85 | 39.10 | 57.28 |\n| MMEB v2 VisDoc | Mean(Task), NDCG@5 | 70.26 | 69.39 | 67.84 |\n| MMEB v2 Video | Mean(Task), Hit@1 | 20.09 | 19.63 | 50.67 |\n| MSEB Retrieval | Mean(Task), MRR@10 | 80.20 | 80.18 | 69.54 |\n| MAEB | Mean(Task) | 58.16 | 58.28 | 49.39 |\n\n| dimensions | storage | MTEB multilingual | MTEB eng | MTEB code | MIEB lite | MMEB overall | MSEB | MAEB |\n| ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |\n| 768 | 1x | 59.55 | 70.25 | 56.99 | 60.93 | 44.19 | 80.20 | 58.16 |\n| 512 | 1/1.5 | 59.20 | 70.24 | 56.62 | 60.77 | 44.70 | 78.99 | 57.93 |\n| 256 | 1/3 | 58.61 | 69.54 | 55.72 | 59.50 | 42.37 | 76.26 | 57.84 |\n| 128 | 1/6 | 55.76 | 67.90 | 52.30 | 55.57 | 37.12 | 60.25 | 55.25 |\n\n| load, one text of 480 to 512 tokens per request | requests/s | p50 (ms) | p99 (ms) |\n| --- | ---: | ---: | ---: |\n| 1 user | 88.5 | 11.4 | 13.1 |\n| 8 users | 207.7 | 38.5 | 40.5 |\n| 32 users | 277.3 | 114.1 | 122.3 |\n\n| single request, warm | latency (ms) |\n| --- | ---: |\n| text, 128 tokens | 5.8 |\n| text, 512 tokens | 7.7 |\n| text, 8192 tokens | 80.3 |\n| image | 47.8 |\n| video, 8 frames | 138.1 |\n| audio, 30 s | 32.2 |\n\nMeasured on one Blackhole chip of a p300c board opened as a 1x1 mesh (equal to a p150, compute grid 11x10), precision policy x02. Accuracy: a fixed subset of each benchmark, scored with the same inputs, prompts and code on this package and on the Hugging Face fp32 reference on CPU; \"card\" is the full-suite score from the google/embeddinggemma-2 card and is not comparable to subset means. No benchmark mean differs from HF fp32 by more than 1.0 point at any dimension. Load: a real server with closed-loop clients, 60 s per level, 0 errors. Single request: host staging, trace replay and readback, without tokenization or media decoding.\n",
133
  "limitations": "- Bring-up build, not a Tenstorrent release.\n- One chip per server; no multi-chip mode.\n- At most 8192 tokens per input, counting the prompt and media tokens.\n- A vector depends slightly on the device batch it ran in (cosine down to 0.98 on hard inputs). Send one input per request for repeatable vectors.\n- Against HF fp32, 8 of the evaluation texts and 37 audio clips fall below 0.995 cosine (minimum 0.959); benchmark means still match within 1.0 point.\n- Audio must be 16 kHz WAV and is cut to 30 s. Video arrives as decoded frames, at most 32. Images use the default 280-token budget.\n- Never run in fp16: the base model card warns it gives NaN or degraded vectors.\n- The first boot compiles kernels and captures 51 traces (about 6 minutes); later boots take about 30 s.\n- Validated on one chip of a p300c. Performance on a p150 with a larger compute grid was not measured.\n",
134
  "risks": null,
 
6
  "device_count": 1,
7
  "producer": {
8
  "tt_kernel_version": "0.1.0",
9
+ "created_at": "2026-10-21T15:24:58.977613+00:00",
10
  "hostname": "qb2-120-p06t08"
11
  },
12
  "weights": {
 
28
  "image": {
29
  "registry": "hf",
30
  "repository": "embeddinggemma-2-p150",
31
+ "tag": "tt-model/embeddinggemma-2-p150:0bd59b87edea",
32
+ "digest": "sha256:0bd59b87edea4b92e96656bdead06a361eac00643fecbca5a1a247c55a460fed"
33
  },
34
  "kind": "tt-dit-server",
35
  "runtime": {
 
102
  "import gradio, json, models.autoports.google_embeddinggemma_2.demo.ui as u, models.autoports.google_embeddinggemma_2.demo.index as i; assert gradio.__version__ == '6.30.0', gradio.__version__; m = json.loads((i.LIBRARY_DIR / 'manifest.json').read_text()); missing = [x['file'] for x in m['items'] if x.get('file') and not (i.LIBRARY_DIR / x['file']).is_file()]; assert not missing, missing[:5]"
103
  ],
104
  "built": {
105
+ "image": "tt-model/embeddinggemma-2-p150:0bd59b87edea",
106
+ "repo": "tt-hous/embeddinggemma-2",
107
  "tt_model_version": "0.1.0",
108
+ "created_at": "2026-10-21T15:23:55+00:00",
109
  "tt_metal": {
110
  "sha": "3be67656b28e3902f0366ea8bab03c41e7e5bd6c",
111
  "describe": null,
 
117
  "pushed": null
118
  },
119
  "code_sha256": "852ada085de3df458409d2183b853e843234bd60acff1efb98eb3c32c62dd77d",
120
+ "image_digest": "sha256:0bd59b87edea4b92e96656bdead06a361eac00643fecbca5a1a247c55a460fed"
121
  },
122
  "card": {
123
  "description": "EmbeddingGemma 2, a multimodal embedding model. Text, code, images, video frames and 16 kHz audio each map to one L2-normalized 768-dimensional vector in a shared space, so a text query can find an image, a sound or a video. Matryoshka dimensions 512, 256 and 128 are the first values of the 768-dimensional vector, re-normalized on the chip.\n\nThe text tower is a bidirectional 24-layer Gemma encoder (hidden size 512, sliding and full attention, 8192-token context). Gemma 4 vision (16 layers) and audio (12 layers) towers feed soft tokens into it. Weights stay on the chip and every request shape runs as a captured TTNN trace. Two custom kernels (sliding-window SDPA and a GEGLU matmul) run through stock `ttnn.generic_op`; tt-metal itself is unmodified.\n",
124
  "attribution": "Derived from google/embeddinggemma-2 by Google DeepMind, weights unmodified. The TTNN port is models/autoports/google_embeddinggemma_2, on stock upstream tt-metal.\n",
125
  "prerequisites": "- 1 GB hugepages mounted at `/dev/hugepages-1G`\n- One chip of a p300 also works: the package sets the p150 mesh graph descriptor.\n",
126
+ "quickstart": "### Embedding text\n```bash\ncurl -s localhost:20000/v1/embeddings -H 'content-type: application/json' \\\n -d '{\"input\": [\"What is the capital of France?\"], \"input_type\": \"query\", \"dimensions\": 256}'\n```\n\n### Demo\n```bash\ntt serve tt-hous/embeddinggemma-2 --profile demo\n```\nOpen `http://<host>:20000/demo/`. A web page where every query is embedded live on the chip, over a bundled library of 163 openly licensed images, videos, sounds, texts in 10 languages and code snippets: media search, finding a moment in a long video, audio search, code search, multilingual search, and a side-by-side Matryoshka comparison.\n",
127
  "architecture": null,
128
  "status": null,
129
  "intended_use": "Embeddings for retrieval, semantic search, clustering and classification over text, code, images, video and audio, served from one p150. Not for generating text.\n",
130
  "out_of_scope_use": null,
131
+ "usage": "`POST /v1/embeddings` follows the OpenAI embeddings request and response shapes, with media extensions.\n\n| field | meaning | default |\n|---|---|---|\n| `input` | a string, or a list of 1 to 2048 strings and media items | required |\n| `dimensions` | 768, 512, 256 or 128 | 768 |\n| `input_type` | a sentence-transformers prompt name, for example `query`, `document`, `CodeRetrieval` | no prompt |\n| `encoding_format` | `float`, or `base64` (little-endian float32) | `float` |\n| `model` | optional; `google/embeddinggemma-2` or `embeddinggemma-2` | |\n\nA media item is an object with optional `text` and at most one media field:\n\n| field | value |\n|---|---|\n| `image` | base64 image bytes (any PIL format) or a `data:image/...;base64,` URL |\n| `audio` | base64 16 kHz WAV; clips over 30 s are cut to 30 s |\n| `video` | `{\"frames\": [<base64 image>, ...], \"fps\": 1}`; at most 32 frames are used |\n\n```bash\nIMG=$(base64 -w0 cats.jpg)\ncurl -s localhost:20000/v1/embeddings -H 'content-type: application/json' \\\n -d \"{\\\"input\\\": [{\\\"image\\\": \\\"$IMG\\\"}, {\\\"text\\\": \\\"two cats on a couch\\\"}]}\"\n```\n\n`GET /health` answers 200 once the traces are captured, with the precision policy and live counters, and 503 before that. `GET /v1/models` lists the model, its dimensions and modalities. Errors use the OpenAI error shape; an input over 8192 tokens gets 400 `context_length_exceeded`. Media URLs are refused: send media as base64.\n",
132
  "performance": "| benchmark (768d) | metric | this package | HF fp32 | card |\n| --- | --- | ---: | ---: | ---: |\n| MTEB multilingual v2 | Mean(Task) | 59.55 | 59.51 | 61.36 |\n| MTEB eng v2 | Mean(Task) | 70.25 | 70.35 | 68.46 |\n| MTEB code v1 | Mean(Task), NDCG@10 | 56.99 | 57.04 | 78.68 |\n| MIEB lite | Mean(TaskType) | 60.93 | 60.50 | 64.64 |\n| MMEB v2 Image | Mean(Task), Hit@1 | 38.85 | 39.10 | 57.28 |\n| MMEB v2 VisDoc | Mean(Task), NDCG@5 | 70.26 | 69.39 | 67.84 |\n| MMEB v2 Video | Mean(Task), Hit@1 | 20.09 | 19.63 | 50.67 |\n| MSEB Retrieval | Mean(Task), MRR@10 | 80.20 | 80.18 | 69.54 |\n| MAEB | Mean(Task) | 58.16 | 58.28 | 49.39 |\n\n| dimensions | storage | MTEB multilingual | MTEB eng | MTEB code | MIEB lite | MMEB overall | MSEB | MAEB |\n| ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |\n| 768 | 1x | 59.55 | 70.25 | 56.99 | 60.93 | 44.19 | 80.20 | 58.16 |\n| 512 | 1/1.5 | 59.20 | 70.24 | 56.62 | 60.77 | 44.70 | 78.99 | 57.93 |\n| 256 | 1/3 | 58.61 | 69.54 | 55.72 | 59.50 | 42.37 | 76.26 | 57.84 |\n| 128 | 1/6 | 55.76 | 67.90 | 52.30 | 55.57 | 37.12 | 60.25 | 55.25 |\n\n| load, one text of 480 to 512 tokens per request | requests/s | p50 (ms) | p99 (ms) |\n| --- | ---: | ---: | ---: |\n| 1 user | 88.5 | 11.4 | 13.1 |\n| 8 users | 207.7 | 38.5 | 40.5 |\n| 32 users | 277.3 | 114.1 | 122.3 |\n\n| single request, warm | latency (ms) |\n| --- | ---: |\n| text, 128 tokens | 5.8 |\n| text, 512 tokens | 7.7 |\n| text, 8192 tokens | 80.3 |\n| image | 47.8 |\n| video, 8 frames | 138.1 |\n| audio, 30 s | 32.2 |\n\nMeasured on one Blackhole chip of a p300c board opened as a 1x1 mesh (equal to a p150, compute grid 11x10), precision policy x02. Accuracy: a fixed subset of each benchmark, scored with the same inputs, prompts and code on this package and on the Hugging Face fp32 reference on CPU; \"card\" is the full-suite score from the google/embeddinggemma-2 card and is not comparable to subset means. No benchmark mean differs from HF fp32 by more than 1.0 point at any dimension. Load: a real server with closed-loop clients, 60 s per level, 0 errors. Single request: host staging, trace replay and readback, without tokenization or media decoding.\n",
133
  "limitations": "- Bring-up build, not a Tenstorrent release.\n- One chip per server; no multi-chip mode.\n- At most 8192 tokens per input, counting the prompt and media tokens.\n- A vector depends slightly on the device batch it ran in (cosine down to 0.98 on hard inputs). Send one input per request for repeatable vectors.\n- Against HF fp32, 8 of the evaluation texts and 37 audio clips fall below 0.995 cosine (minimum 0.959); benchmark means still match within 1.0 point.\n- Audio must be 16 kHz WAV and is cut to 30 s. Video arrives as decoded frames, at most 32. Images use the default 280-token budget.\n- Never run in fp16: the base model card warns it gives NaN or degraded vectors.\n- The first boot compiles kernels and captures 51 traces (about 6 minutes); later boots take about 30 s.\n- Validated on one chip of a p300c. Performance on a p150 with a larger compute grid was not measured.\n",
134
  "risks": null,