Bump the pip group across 3 directories with 5 updates

[dependabot]

Bumps the pip group with 5 updates in the / directory:

Package	From	To
requests	2.32.3	2.32.4
torch	2.5.0	2.8.0
transformers	4.45.2	4.53.0
vllm	0.6.3	0.11.0
urllib3	1.26.6	2.5.0

Bumps the pip group with 1 update in the /docs directory: urllib3.
Bumps the pip group with 3 updates in the /rankify directory: requests, torch and transformers.

Updates requests from 2.32.3 to 2.32.4

Release notes

Sourced from requests's releases.

v2.32.4

2.32.4 (2025-06-10)

Security

• CVE-2024-47081 Fixed an issue where a maliciously crafted URL and trusted environment will retrieve credentials for the wrong hostname/machine from a netrc file. (#6965)

Improvements

• Numerous documentation improvements

Deprecations

• Added support for pypy 3.11 for Linux and macOS. (#6926)
• Dropped support for pypy 3.9 following its end of support. (#6926)

Changelog

Sourced from requests's changelog.

2.32.4 (2025-06-10)

Security

• CVE-2024-47081 Fixed an issue where a maliciously crafted URL and trusted environment will retrieve credentials for the wrong hostname/machine from a netrc file.

Improvements

• Numerous documentation improvements

Deprecations

• Added support for pypy 3.11 for Linux and macOS.
• Dropped support for pypy 3.9 following its end of support.

Commits

• 021dc72 Polish up release tooling for last manual release
• 821770e Bump version and add release notes for v2.32.4
• 59f8aa2 Add netrc file search information to authentication documentation (#6876)
• 5b4b64c Add more tests to prevent regression of CVE 2024 47081
• 7bc4587 Add new test to check netrc auth leak (#6962)
• 96ba401 Only use hostname to do netrc lookup instead of netloc
• 7341690 Merge pull request #6951 from tswast/patch-1
• 6716d7c remove links
• a7e1c74 Update docs/conf.py
• c799b81 docs: fix dead links to kenreitz.org
• Additional commits viewable in compare view

Updates torch from 2.5.0 to 2.8.0

Release notes

Sourced from torch's releases.

PyTorch 2.8.0 Release Notes

• Highlights
• Backwards Incompatible Changes
• Deprecations
• New Features
• Improvements
• Bug fixes
• Performance
• Documentation
• Developers

Highlights

... (truncated)

Commits

• ba56102 Cherrypick: Add the RunLLM widget to the website (#159592)
• c525a02 [dynamo, docs] cherry pick torch.compile programming model docs into 2.8 (#15...
• a1cb3cc [Release Only] Remove nvshmem from list of preload libraries (#158925)
• c76b235 Move out super large one off foreach_copy test (#158880)
• 20a0e22 Revert "[Dynamo] Allow inlining into AO quantization modules (#152934)" (#158...
• 9167ac8 [MPS] Switch Cholesky decomp to column wise (#158237)
• 5534685 [MPS] Reimplement tri[ul] as Metal shaders (#158867)
• d19e08d Cherry pick PR 158746 (#158801)
• a6c044a [cherry-pick] Unify torch.tensor and torch.ops.aten.scalar_tensor behavior (#...
• 620ebd0 [Dynamo] Use proper sources for constructing dataclass defaults (#158689)
• Additional commits viewable in compare view

Updates transformers from 4.45.2 to 4.53.0

Release notes

Sourced from transformers's releases.

Release v4.53.0

Gemma3n

Gemma 3n models are designed for efficient execution on low-resource devices. They are capable of multimodal input, handling text, image, video, and audio input, and generating text outputs, with open weights for pre-trained and instruction-tuned variants. These models were trained with data in over 140 spoken languages.

Gemma 3n models use selective parameter activation technology to reduce resource requirements. This technique allows the models to operate at an effective size of 2B and 4B parameters, which is lower than the total number of parameters they contain. For more information on Gemma 3n's efficient parameter management technology, see the Gemma 3n page.

from transformers import pipeline
import torch
pipe = pipeline(
"image-text-to-text",
torch_dtype=torch.bfloat16,
model="google/gemma-3n-e4b",
device="cuda",
)
output = pipe(
"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg",
text="<image_soft_token> in this image, there is"
)
print(output)

Dia

Dia is an opensource text-to-speech (TTS) model (1.6B parameters) developed by Nari Labs. It can generate highly realistic dialogue from transcript including nonverbal communications such as laughter and coughing. Furthermore, emotion and tone control is also possible via audio conditioning (voice cloning).

Model Architecture: Dia is an encoder-decoder transformer based on the original transformer architecture. However, some more modern features such as rotational positional embeddings (RoPE) are also included. For its text portion (encoder), a byte tokenizer is utilized while for the audio portion (decoder), a pretrained codec model DAC is used - DAC encodes speech into discrete codebook tokens and decodes them back into audio.

• Add Dia model by @​buttercrab in #38405

Kyutai Speech-to-Text

Kyutai STT is a speech-to-text model architecture based on the Mimi codec, which encodes audio into discrete tokens in a streaming fashion, and a Moshi-like autoregressive decoder. Kyutai’s lab has released two model checkpoints:

• kyutai/stt-1b-en_fr: a 1B-parameter model capable of transcribing both English and French

... (truncated)

Commits

• 67ddc82 Release: v4.53.0
• 0a8081b [Modeling] Fix encoder CPU offloading for whisper (#38994)
• c63cfd6 Gemma 3n (#39059)
• 3e5cc12 [tests] remove tests from libraries with deprecated support (flax, tensorflow...
• cfff7ca [Whisper] Pipeline: handle long form generation (#35750)
• 02ecdcf add _keep_in_fp32_modules_strict (#39058)
• d973e62 fix condition where torch_dtype auto collides with model_kwargs. (#39054)
• 44b2316 [qwen2-vl] fix vision attention scaling (#39043)
• ae15715 polishing docs: error fixes for clarity (#39042)
• 3abeaba Create test for #38916 (custom generate from local dir with imports) (#39015)
• Additional commits viewable in compare view

Updates vllm from 0.6.3 to 0.11.0

Release notes

Sourced from vllm's releases.

v0.11.0

Highlights

This release features 538 commits, 207 contributors (65 new contributors)!

• This release completes the removal of V0 engine. V0 engine code including AsyncLLMEngine, LLMEngine, MQLLMEngine, all attention backends, and related components have been removed. V1 is the only engine in the codebase now.
• This releases turns on FULL_AND_PIECEWISE as the CUDA graph mode default. This should provide better out of the box performance for most models, particularly fine-grained MoEs, while preserving compatibility with existing models supporting only PIECEWISE mode.

Note: In v0.11.0 (and v0.10.2), --async-scheduling will produce gibberish output in some cases such as preemption and others. This functionality is correct in v0.10.1. We are actively fixing it for the next version.

Model Support

• New architectures: DeepSeek-V3.2-Exp (#25896), Qwen3-VL series (#24727), Qwen3-Next (#24526), OLMo3 (#24534), LongCat-Flash (#23991), Dots OCR (#24645), Ling2.0 (#24627), CWM (#25611).
• Encoders: RADIO encoder support (#24595), Transformers backend support for encoder-only models (#25174).
• Task expansion: BERT token classification/NER (#24872), multimodal models for pooling tasks (#24451).
• Data parallel for vision encoders: InternVL (#23909), Qwen2-VL (#25445), Qwen3-VL (#24955).
• Speculative decoding: EAGLE3 for MiniCPM3 (#24243) and GPT-OSS (#25246).
• Features: Qwen3-VL text-only mode (#26000), EVS video token pruning (#22980), Mamba2 TP+quantization (#24593), MRoPE + YaRN (#25384), Whisper on XPU (#25123), LongCat-Flash-Chat tool calling (#24083).
• Performance: GLM-4.1V 916ms TTFT reduction via fused RMSNorm (#24733), GLM-4 MoE SharedFusedMoE optimization (#24849), Qwen2.5-VL CUDA sync removal (#24741), Qwen3-VL Triton MRoPE kernel (#25055), FP8 checkpoints for Qwen3-Next (#25079).
• Reasoning: SeedOSS reason parser (#24263).

Engine Core

• KV cache offloading: CPU offloading with LRU management (#19848, #20075, #21448, #22595, #24251).
• V1 features: Prompt embeddings (#24278), sharded state loading (#25308), FlexAttention sliding window (#24089), LLM.apply_model (#18465).
• Hybrid allocator: Pipeline parallel (#23974), varying hidden sizes (#25101).
• Async scheduling: Uniprocessor executor support (#24219).
• Architecture: Tokenizer group removal (#24078), shared memory multimodal caching (#20452).
• Attention: Hybrid SSM/Attention in Triton (#21197), FlashAttention 3 for ViT (#24347).
• Performance: FlashInfer RoPE 2x speedup (#21126), fused Q/K RoPE 11% improvement (#24511, #25005), 8x spec decode overhead reduction (#24986), FlashInfer spec decode with 1.14x speedup (#25196), model info caching (#23558), inputs_embeds copy avoidance (#25739).
• LoRA: Optimized weight loading (#25403).
• Defaults: CUDA graph mode FULL_AND_PIECEWISE (#25444), Inductor standalone compile disabled (#25391).
• torch.compile: CUDA graph Inductor partition integration (#24281).

Hardware & Performance

• NVIDIA: FP8 FlashInfer MLA decode (#24705), BF16 fused MoE for Hopper/Blackwell expert parallel (#25503).
• DeepGEMM: Enabled by default (#24462), 5.5% throughput improvement (#24783).
• New architectures: RISC-V 64-bit (#22112), ARM non-x86 CPU (#25166), ARM 4-bit fused MoE (#23809).
• AMD: ROCm 7.0 (#25178), GLM-4.5 MI300X tuning (#25703).
• Intel XPU: MoE DP accuracy fix (#25465).

Large Scale Serving & Performance

• Dual-Batch Overlap (DBO): Overlapping computation mechanism (#23693), DeepEP high throughput + prefill (#24845).
• Data Parallelism: torchrun launcher (#24899), Ray placement groups (#25026), Triton DP/EP kernels (#24588).
• EPLB: Hunyuan V1 (#23078), Mixtral (#22842), static placement (#23745), reduced overhead (#24573).
• Disaggregated serving: KV transfer metrics (#22188), NIXL MLA latent dimension (#25902).
• MoE: Shared expert overlap optimization (#24254), SiLU kernel for DeepSeek-R1 (#24054), Enable Allgather/ReduceScatter backend for NaiveAllToAll (#23964).
• Distributed: NCCL symmetric memory with 3-4% throughput improvement (#24532), enabled by default for TP (#25070).

Quantization

• FP8: Per-token-group quantization (#24342), hardware-accelerated instructions (#24757), torch.compile KV cache (#22758), paged attention update (#22222).
• FP4: NVFP4 for dense models (#25609), Gemma3 (#22771), Llama 3.1 405B (#25135).
• W4A8: Faster preprocessing (#23972).

... (truncated)

Commits

• b8b302c Update CUDA architecture list in build pipeline for 12.9.1 wheels (#26592)
• f71952c [Build/CI] Revert back to Ubuntu 20.04, install python 3.12 with uv (#26103)
• d100776 [Bugfix] Disable cascade attention with FlashInfer (#26130)
• c75c2e7 [Deepseek v3.2] Support indexer prefill chunking (#25999)
• 9d9a2b7 [Small] Prevent bypassing media domain restriction via HTTP redirects (#26035)
• 6040e0b [BugFix] Fix FI accuracy issue when used for MLA prefill (#26063)
• 05bf0c5 Update base image to 22.04 (jammy) (#26065)
• c536881 [BugFix] ChunkedLocalAttention is currently not CG compatible (#26034)
• ebce361 [BugFix][DP/EP] Fix CUTLASS MLA hang under load (#26026)
• e4beabd [BugFix] Fix default kv-cache-dtype default for DeepseekV3.2 (#25988)
• Additional commits viewable in compare view

Updates urllib3 from 1.26.6 to 2.5.0

Release notes

Sourced from urllib3's releases.

2.5.0

🚀 urllib3 is fundraising for HTTP/2 support

urllib3 is raising ~$40,000 USD to release HTTP/2 support and ensure long-term sustainable maintenance of the project after a sharp decline in financial support. If your company or organization uses Python and would benefit from HTTP/2 support in Requests, pip, cloud SDKs, and thousands of other projects please consider contributing financially to ensure HTTP/2 support is developed sustainably and maintained for the long-haul.

Thank you for your support.

Security issues

urllib3 2.5.0 fixes two moderate security issues:

• Pool managers now properly control redirects when retries is passed — CVE-2025-50181 reported by @​sandumjacob (5.3 Medium, GHSA-pq67-6m6q-mj2v)
• Redirects are now controlled by urllib3 in the Node.js runtime — CVE-2025-50182 (5.3 Medium, GHSA-48p4-8xcf-vxj5)

Features

• Added support for the compression.zstd module that is new in Python 3.14. See PEP 784 for more information. (#3610)
• Added support for version 0.5 of hatch-vcs (#3612)

Bugfixes

• Raised exception for HTTPResponse.shutdown on a connection already released to the pool. (#3581)
• Fixed incorrect CONNECT statement when using an IPv6 proxy with connection_from_host. Previously would not be wrapped in []. (#3615)

2.4.0

🚀 urllib3 is fundraising for HTTP/2 support

urllib3 is raising ~$40,000 USD to release HTTP/2 support and ensure long-term sustainable maintenance of the project after a sharp decline in financial support. If your company or organization uses Python and would benefit from HTTP/2 support in Requests, pip, cloud SDKs, and thousands of other projects please consider contributing financially to ensure HTTP/2 support is developed sustainably and maintained for the long-haul.

Thank you for your support.

Features

• Applied PEP 639 by specifying the license fields in pyproject.toml. (#3522)
• Updated exceptions to save and restore more properties during the pickle/serialization process. (#3567)
• Added verify_flags option to create_urllib3_context with a default of VERIFY_X509_PARTIAL_CHAIN and VERIFY_X509_STRICT for Python 3.13+. (#3571)

Bugfixes

• Fixed a bug with partial reads of streaming data in Emscripten. (#3555)

Misc

• Switched to uv for installing development dependecies. (#3550)
• Removed the multiple.intoto.jsonl asset from GitHub releases. Attestation of release files since v2.3.0 can be found on PyPI. (#3566)

2.3.0

... (truncated)

Changelog

Sourced from urllib3's changelog.

2.5.0 (2025-06-18)

Features

• Added support for the compression.zstd module that is new in Python 3.14. See PEP 784 <https://peps.python.org/pep-0784/>_ for more information. ([#3610](https://github.com/urllib3/urllib3/issues/3610) <https://github.com/urllib3/urllib3/issues/3610>__)
• Added support for version 0.5 of hatch-vcs ([#3612](https://github.com/urllib3/urllib3/issues/3612) <https://github.com/urllib3/urllib3/issues/3612>__)

Bugfixes

• Fixed a security issue where restricting the maximum number of followed redirects at the urllib3.PoolManager level via the retries parameter did not work.
• Made the Node.js runtime respect redirect parameters such as retries and redirects.
• Raised exception for HTTPResponse.shutdown on a connection already released to the pool. ([#3581](https://github.com/urllib3/urllib3/issues/3581) <https://github.com/urllib3/urllib3/issues/3581>__)
• Fixed incorrect CONNECT statement when using an IPv6 proxy with connection_from_host. Previously would not be wrapped in []. ([#3615](https://github.com/urllib3/urllib3/issues/3615) <https://github.com/urllib3/urllib3/issues/3615>__)

2.4.0 (2025-04-10)

Features

• Applied PEP 639 by specifying the license fields in pyproject.toml. ([#3522](https://github.com/urllib3/urllib3/issues/3522) <https://github.com/urllib3/urllib3/issues/3522>__)
• Updated exceptions to save and restore more properties during the pickle/serialization process. ([#3567](https://github.com/urllib3/urllib3/issues/3567) <https://github.com/urllib3/urllib3/issues/3567>__)
• Added verify_flags option to create_urllib3_context with a default of VERIFY_X509_PARTIAL_CHAIN and VERIFY_X509_STRICT for Python 3.13+. ([#3571](https://github.com/urllib3/urllib3/issues/3571) <https://github.com/urllib3/urllib3/issues/3571>__)

Bugfixes

• Fixed a bug with partial reads of streaming data in Emscripten. ([#3555](https://github.com/urllib3/urllib3/issues/3555) <https://github.com/urllib3/urllib3/issues/3555>__)

Misc

• Switched to uv for installing development dependecies. ([#3550](https://github.com/urllib3/urllib3/issues/3550) <https://github.com/urllib3/urllib3/issues/3550>__)
• Removed the multiple.intoto.jsonl asset from GitHub releases. Attestation of release files since v2.3.0 can be found on PyPI. ([#3566](https://github.com/urllib3/urllib3/issues/3566) <https://github.com/urllib3/urllib3/issues/3566>__)

2.3.0 (2024-12-22)

... (truncated)

Commits

• aaab4ec Release 2.5.0
• 7eb4a2a Merge commit from fork
• f05b132 Merge commit from fork
• d03fe32 Fix HTTP tunneling with IPv6 in older Python versions
• 11661e9 Bump github/codeql-action from 3.28.0 to 3.29.0 (#3624)
• 6a0ecc6 Update v2 migration guide to 2.4.0 (#3621)
• 8e32e60 Raise exception for shutdown on a connection already released to the pool (#3...
• 9996e0f Fix emscripten CI for Chrome 137+ (#3599)
• 4fd1a99 Bump RECENT_DATE (#3617)
• c4b5917 Add support for the new compression.zstd module in Python 3.14 (#3611)
• Additional commits viewable in compare view

Updates urllib3 from 1.26.6 to 2.5.0

Release notes

Sourced from urllib3's releases.

2.5.0

🚀 urllib3 is fundraising for HTTP/2 support

urllib3 is raising ~$40,000 USD to release HTTP/2 support and ensure long-term sustainable maintenance of the project after a sharp decline in financial support. If your company or organization uses Python and would benefit from HTTP/2 support in Requests, pip, cloud SDKs, and thousands of other projects please consider contributing financially to ensure HTTP/2 support is developed sustainably and maintained for the long-haul.

Thank you for your support.

Security issues

urllib3 2.5.0 fixes two moderate security issues:

• Pool managers now properly control redirects when retries is passed — CVE-2025-50181 reported by @​sandumjacob (5.3 Medium, GHSA-pq67-6m6q-mj2v)
• Redirects are now controlled by urllib3 in the Node.js runtime — CVE-2025-50182 (5.3 Medium, GHSA-48p4-8xcf-vxj5)

Features

• Added support for the compression.zstd module that is new in Python 3.14. See PEP 784 for more information. (#3610)
• Added support for version 0.5 of hatch-vcs (#3612)

Bugfixes

• Raised exception for HTTPResponse.shutdown on a connection already released to the pool. (#3581)
• Fixed incorrect CONNECT statement when using an IPv6 proxy with connection_from_host. Previously would not be wrapped in []. (#3615)

2.4.0

🚀 urllib3 is fundraising for HTTP/2 support

urllib3 is raising ~$40,000 USD to release HTTP/2 support and ensure long-term sustainable maintenance of the project after a sharp decline in financial support. If your company or organization uses Python and would benefit from HTTP/2 support in Requests, pip, cloud SDKs, and thousands of other projects please consider contributing financially to ensure HTTP/2 support is developed sustainably and maintained for the long-haul.

Thank you for your support.

Features

• Applied PEP 639 by specifying the license fields in pyproject.toml. (#3522)
• Updated exceptions to save and restore more properties during the pickle/serialization process. (#3567)
• Added verify_flags option to create_urllib3_context with a default of VERIFY_X509_PARTIAL_CHAIN and VERIFY_X509_STRICT for Python 3.13+. (#3571)

Bugfixes

• Fixed a bug with partial reads of streaming data in Emscripten. (#3555)

Misc

• Switched to uv for installing development dependecies. (#3550)
• Removed the multiple.intoto.jsonl asset from GitHub releases. Attestation of release files since v2.3.0 can be found on PyPI. (#3566)

2.3.0

... (truncated)

Changelog

Sourced from urllib3's changelog.

2.5.0 (2025-06-18)

Features

• Added support for the compression.zstd module that is new in Python 3.14. See PEP 784 <https://peps.python.org/pep-0784/>_ for more information. ([#3610](https://github.com/urllib3/urllib3/issues/3610) <https://github.com/urllib3/urllib3/issues/3610>__)
• Added support for version 0.5 of hatch-vcs ([#3612](https://github.com/urllib3/urllib3/issues/3612) <https://github.com/urllib3/urllib3/issues/3612>__)

Bugfixes

• Fixed a security issue where restricting the maximum number of followed redirects at the urllib3.PoolManager level via the retries parameter did not work.
• Made the Node.js runtime respect redirect parameters such as retries and redirects.
• Raised exception for HTTPResponse.shutdown on a connection already released to the pool. ([#3581](https://github.com/urllib3/urllib3/issues/3581) <https://github.com/urllib3/urllib3/issues/3581>__)
• Fixed incorrect CONNECT statement when using an IPv6 proxy with connection_from_host. Previously would not be wrapped in []. ([#3615](https://github.com/urllib3/urllib3/issues/3615) <https://github.com/urllib3/urllib3/issues/3615>__)

2.4.0 (2025-04-10)

Features

• Applied PEP 639 by specifying the license fields in pyproject.toml. ([#3522](https://github.com/urllib3/urllib3/issues/3522) <https://github.com/urllib3/urllib3/issues/3522>__)
• Updated exceptions to save and restore more properties during the pickle/serialization process. ([#3567](https://github.com/urllib3/urllib3/issues/3567) <https://github.com/urllib3/urllib3/issues/3567>__)
• Added verify_flags option to create_urllib3_context with a default of VERIFY_X509_PARTIAL_CHAIN and VERIFY_X509_STRICT for Python 3.13+. ([#3571](https://github.com/urllib3/urllib3/issues/3571) <https://github.com/urllib3/urllib3/issues/3571>__)

Bugfixes

• Fixed a bug with partial reads of streaming data in Emscripten. ([#3555](https://github.com/urllib3/urllib3/issues/3555) <https://github.com/urllib3/urllib3/issues/3555>__)

Misc

• Switched to uv for installing development dependecies. ([#3550](https://github.com/urllib3/urllib3/issues/3550) <https://github.com/urllib3/urllib3/issues/3550>__)
• Removed the multiple.intoto.jsonl asset from GitHub releases. Attestation of release files since v2.3.0 can be found on PyPI. ([#3566](https://github.com/urllib3/urllib3/issues/3566) <https://github.com/urllib3/urllib3/issues/3566>__)

2.3.0 (2024-12-22)

... (truncated)

Commits

• aaab4ec Release 2.5.0
• 7eb4a2a Merge commit from fork
• f05b132 Merge commit from fork
• d03fe32 Fix HTTP tunneling with IPv6 in older Python versions
• 11661e9 Bump github/codeql-action from 3.28.0 to 3.29.0 (#3624)
• 6a0ecc6 Update v2 migration guide to 2.4.0 (#3621)
• 8e32e60 Raise exception for shutdown on a connection already released to the pool (#3...
• 9996e0f Fix emscripten CI for Chrome 137+ (#3599)
• 4fd1a99 Bump RECENT_DATE (#3617)
• c4b5917 Add support for the new compression.zstd module in Python 3.14 (#3611)
• Additional commits viewable in compare view

Updates requests from 2.32.3 to 2.32.4

Release notes

Sourced from requests's releases.

v2.32.4

2.32.4 (2025-06-10)

Security

• CVE-2024-47081 Fixed an issue where a maliciously crafted URL and trusted environment will retrieve credentials for the wrong hostname/machine from a netrc file. (#6965)

Improvements

• Numerous documentation improvements

Deprecations

• Added support for pypy 3.11 for Linux and macOS. (#6926)
• Dropped support for pypy 3.9 following its end of support. (#6926)

Changelog

Sourced from requests's changelog.

2.32.4 (2025-06-10)

Security

• CVE-2024-47081 Fixed an issue where a maliciously crafted URL and trusted environment will retrieve credentials for the wrong hostname/machine from a netrc file.

Improvements

• Numerous documentation improvements

Deprecations

• Added support for pypy 3.11 for Linux and macOS.
• Dropped support for pypy 3.9 following its end of support.

Commits

• 021dc72 Polish up release tooling for last manual release
• 821770e Bump version and add release notes for v2.32.4
• 59f8aa2 Add netrc file search information to authentication documentation (#6876)
• 5b4b64c Add more tests to prevent regression of CVE 2024 47081
• 7bc4587 Add new test to check netrc auth leak (#6962)
• 96ba401 Only use hostname to do netrc lookup instead of netloc
• 7341690 Merge pull request #6951 from tswast/patch-1
• 6716d7c remove links
• a7e1c74 Update docs/conf.py
• c799b81 docs: fix dead links to kenreitz.org
• Additional commits viewable in compare view

Updates torch from 2.5.0 to 2.8.0

Release notes

Sourced from torch's releases.

PyTorch 2.8.0 Release Notes

• Highlights
• Backwards Incompatible Changes
• Deprecations
• New Features
• Improvements
• Bug fixes
• Performance
• Documentation
• Developers

Highlights

... (truncated)

Commits

• ba56102 Cherrypick: Add the RunLLM widget to the website (#159592)
• c525a02 [dynamo, docs] cherry pick torch.compile programming model docs into 2.8 (#15...
• a1cb3cc [Release Only] Remove nvshmem from list of preload libraries (#158925)
• c76b235 Move out super large one off foreach_copy test (#158880)
• 20a0e22 Revert "[Dynamo] Allow inlining into AO quantization modules (#152934)" (#158...
• 9167ac8 [MPS] Switch Cholesky decomp to column wise (#158237)
• 5534685 [MPS] Reimplement tri[ul] as Metal shaders (#158867)
• d19e08d Cherry pick PR 158746 (#158801)
• a6c044a [cherry-pick] Unify torch.tensor and torch.ops.aten.scalar_tensor behavior (#...
• 620ebd0 [Dynamo] Use proper sources for constructing dataclass defaults (#158689)
• Additional commits viewable in compare view

Updates transformers from 4.45.2 to 4.53.0

Release notes

Sourced from transformers's releases.

Release v4.53.0

Gemma3n

Gemma 3n models are designed for efficient execution on low-resource devices. They are capable of multimodal input, handling text, image, video, and audio input, and generating text outputs, with open weights for pre-trained and instruction-tuned variants. These models were trained with data in over 140 spoken languages.

Gemma 3n models use selective parameter activation technology to reduce resource requirements. This technique allows the models to operate at an effective size of 2B and 4B parameters, which is lower than the total number of parameters they contain. For more information on Gemma 3n's efficient parameter management technology, see the Gemma 3n page.

from transformers import pipeline
import torch
pipe = pipeline(
"image-text-to-text",
torch_dtype=torch.bfloat16,
model="google/gemma-3n-e4b",
device="cuda",
)
output = pipe(
"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg",
text="<image_soft_token> in this image, there is"
)
print(output)

Dia

Dia is an opensource text-to-speech (TTS) model (1.6B parameters) developed by Nari Labs. It can generate highly realistic dialogue from transcript including nonverbal communications such as laughter and coughing. Furthermore, emotion and tone control is also possible via audio conditioning (voice cloning).

Model Architecture: Dia is an encoder-decoder transformer based on the original transformer architecture. However, some more modern features such as rotational positional embeddings (RoPE) are also included. For its text portion (encoder), a byte tokenizer is utilized while for the audio portion (decoder), a pretrained codec model DAC is used - DAC encodes speech into discrete codebook tokens and decodes them back into audio.

• Add Dia model by @​buttercrab in #38405

Kyutai Speech-to-Text

Kyutai STT is a speech-to-text model architecture based on the Mimi codec, which encodes audio into discrete tokens in a streaming fashion, and a Moshi-like autoregressive decoder. Kyutai’s lab has released two model checkpoints:

• kyutai/stt-1b-en_fr: a 1B-parameter model capable of transcribing both English and French

... (truncated)

Commits

• 67ddc82 Release: v4.53.0
• 0a8081b [Modeling] Fix encoder CPU offloading for whisper (#38994)
• c63cfd6 Gemma 3n (#39059)
• 3e5cc12 [tests] remove tests from libraries with deprecated support (flax, tensorflow...
• cfff7ca [Whisper] Pipeline: handle long form generation (#35750)
• 02ecdcf add _keep_in_fp32_modules_strict (#39058)
• d973e62 fix condition where torch_dtype auto collides with model_kwargs. (#39054)
• 44b2316 [qwen2-vl] fix vision attention scaling (#39043)
• ae15715 polishing docs: error fixes for clarity (#39042)
• 3abeaba Create test for #38916 (custom generate from local dir with imports) (#39015)
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions
  You can disable automated security fix PRs for this repo from the Security Alerts page.