chore(deps): bump the uv group across 1 directory with 3 updates

[dependabot]

Bumps the uv group with 3 updates in the / directory: sentence-transformers, anyio and urllib3.

Updates sentence-transformers from 5.5.1 to 5.6.0

Release notes

Sourced from sentence-transformers's releases.

v5.6.0 - Fixes for Causal LM Rerankers, Hard-Negative Mining, and More

This minor version is a correctness- and robustness-focused release. It fixes a silent scoring bug for causal-LM rerankers, corrects several hard-negative mining and GIST loss edge cases, restores TSDAE on transformers v5, and adds Apple Silicon (MPS) support for the cached losses.

The headline fix affects chat-template models that read the final token position, i.e. causal-LM rerankers (like Qwen3-Reranker) and last-token-pooling embedders: when an over-long input was truncated, the chat template's trailing suffix (e.g. the assistant prefill the model scores from) was silently dropped, producing wrong scores with no error. There's also a forward-looking deprecation: loading local custom code without trust_remote_code=True now warns, and will require it from v6.0.

Install this version with

# Training + Inference
pip install sentence-transformers[train]==5.6.0
Inference only, use one of:
pip install sentence-transformers==5.6.0
pip install sentence-transformers[onnx-gpu]==5.6.0
pip install sentence-transformers[onnx]==5.6.0
pip install sentence-transformers[openvino]==5.6.0
Multimodal dependencies (optional):
pip install sentence-transformers[image]==5.6.0
pip install sentence-transformers[audio]==5.6.0
pip install sentence-transformers[video]==5.6.0
Or combine as needed:
pip install sentence-transformers[train,onnx,image]==5.6.0

Fixed silently wrong scores when truncation drops chat-template suffixes (#3787)

Chat-template models render the full conversation to a flat string before tokenizing, so when the rendered input is longer than the tokenizer's model_max_length, the tokenizer truncates it from the right and drops the template's trailing suffix: the fixed tokens a template appends after the content, e.g. a prompt, instruction, [/INST], or a trailing EOS. For models that read the final token position, this silently corrupted the result:

• causal-LM rerankers (e.g. Qwen/Qwen3-Reranker-0.6B) score a pair from the last token's yes/no logits, and
• last-token-pooling embedders read the final hidden state.

When the suffix was truncated away, that final position landed mid-document instead of after the prefill, so the score or embedding came from the wrong place.

Transformer.preprocess now detects when truncation drops the suffix and splices it back onto the tail of each truncated row. Because the fix lives in the shared base Transformer, it applies across SentenceTransformer, CrossEncoder, and SparseEncoder. It's enabled by default and saved to the model configuration. Pass processing_kwargs={"chat_template": {"restore_suffix": False}} to opt back into raw truncation.

Hard-negative mining and GIST loss correctness (#3821, #3817, #3816)

A trio of correctness and scalability fixes for hard-negative mining and the GIST losses:

• Sign-independent relative margin: mine_hard_negatives(relative_margin=...) and the margin_strategy="relative" branch of GISTEmbedLoss / CachedGISTEmbedLoss used a multiplicative threshold (positive * (1 - margin)) that only behaves correctly when the positive-pair similarity is positive. When that similarity was negative, the threshold moved the wrong way and let through false negatives: candidates more similar to the anchor than the true positive. The threshold is now positive - |positive| * margin, identical to before for positive scores but correct for negative ones.
• Distributed positive masking in the GIST losses: with gather_across_devices=True and a non-zero margin, the false-negative suppression mask protected the wrong columns on ranks beyond the first (it ignored the per-rank offset into the gathered batch), which set the true positive's logit to -inf and produced a +inf loss. The mask now accounts for the cross-rank offset, so multi-GPU GIST training stays finite.
• Memory-bounded mining without FAISS: mine_hard_negatives(use_faiss=False) (the default) materialized the full (queries × corpus) similarity matrix at once, which could OOM on large corpora. It now batches over the query axis (controlled by faiss_batch_size, default 16384), bounding peak memory while producing identical results.

TSDAE weight tying restored on transformers v5 (#3781)

transformers v5 removed the private PreTrainedModel._tie_encoder_decoder_weights helper that DenoisingAutoEncoderLoss (TSDAE) used to tie its separate encoder and decoder. As a stopgap, v5.5 raised a RuntimeError for the default tie_encoder_decoder=True on transformers >= 5.0.0, effectively breaking TSDAE there unless you pinned an older transformers or disabled tying. TSDAE now ships its own tying routine that shares storage between encoder and decoder, so it works on both transformers <5 and >=5 with the default settings.

Deprecation: loading local custom code without trust_remote_code (#3807)

... (truncated)

Commits

• 9c73df3 Release v5.6.0
• 222b052 [fix] Don't override device_map placement with the device argument (#3823)
• a38a6bf Fix causal LM reranker scoring when max_length truncates chat-template suffix...
• 1812103 [fix] Make relative margin sign-independent in mining and GIST losses (#3821)
• ae1acc3 Warn when loading local custom code without trust_remote_code (#3807)
• 429cf5d [fix] Support MPS in the cached losses' RandContext (#3812)
• 77fdbff [fix] fix MPS errors (#3818)
• bfba988 [fix] Fix positive masking in GIST losses with multi-GPU + gather_across_devi...
• 29e382b [docs] Fix doc build problems (part 1) (#3811)
• d16e6bf [fix] Avoid materializing the full similarity matrix in mine_hard_negatives w...
• Additional commits viewable in compare view

Updates anyio from 4.13.0 to 4.14.2

Release notes

Sourced from anyio's releases.

4.14.2

• Changed ByteReceiveStream.receive() implementations to raise a ValueError when max_bytes is not a positive integer (#1191)
• Fixed CapacityLimiter.total_tokens rejecting float("inf") when the limiter was instantiated outside of an event loop. The adapter setter checked for infinity by identity (value is math.inf), so only the exact math.inf singleton was accepted, while every backend setter (using math.isinf()) accepts any positive infinity (#1189; PR by @​greymoth-jp).
• Fixed to_process.run_sync() deadlocking when the worker function writes enough data to sys.stderr to fill the (undrained) pipe buffer. The worker process now redirects sys.stderr to os.devnull as well, matching the documented behavior
• Fixed TLSStream.wrap() matching an internationalized (unicode) host name against the peer certificate using IDNA 2003 (via the standard library) instead of IDNA 2008, which could cause the host name to be matched against the wrong certificate (#1208)
• Fixed anyio.open_process() (and run_process()) ignoring the extra_groups argument, as it mistakenly passed the value of the group argument instead (#1209)
• Fixed CapacityLimiter.acquire_nowait() and CapacityLimiter.acquire_nowait_on_behalf_of() raising trio.WouldBlock instead of anyio.WouldBlock on the trio backend when there are no tokens available (#1218)
• Fixed CapacityLimiter on the asyncio backend over-granting tokens (borrowed_tokens exceeding total_tokens and available_tokens going negative) when a non-blocking acquire was made in the window between a token being released and the notified waiter resuming. The freed token is now reserved for the woken waiter right away, so the non-blocking acquire correctly raises WouldBlock (#1170; PR by @​gaoflow)
• Fixed unnecessary CPU spin when delivering cancellation from CancelScope on asyncio under certain conditions, including improper cancel scope nesting (#1111)

4.14.1

• Fixed teardown of higher-scoped async fixtures failing on asyncio with RuntimeError: Attempted to exit cancel scope in a different task than it was entered in when an async test raise an outcome exception (e.g., pytest.skip(), pytest.xfail(), or pytest.fail()) (#1179; PR by @​EmmanuelNiyonshuti)
• Fixed CapacityLimiter.total_tokens rejecting a value of 0 when the limiter was instantiated outside of an event loop, contradicting the documented behavior of allowing 0 total tokens (#1183; PR by @​nyxst4ck)

4.14.0

• Added support for Python 3.15

• Added an asynchronous implementation of the itertools module (#998; PR by @​11kkw)

• Added the local_port parameter to connect_tcp() to allow binding to a specific local port before connecting (#1067; PR by @​nullwiz)

• Added support for custom capacity limiters in async path and file I/O functions and classes

• Added the create_task() task group method for easier asyncio migration (returns a TaskHandle) (#1098)

• Changed TaskGroup.start_soon() to return a TaskHandle

• Added an option for TaskGroup.start() to return a TaskHandle (which then contains the start value in the start_value property)

• Added the cancel() convenience method to TaskGroup as a shortcut for cancelling the task group's cancel scope

• Improved the error message when a known backend is not installed to suggest the install command (#1115; PR by @​EmmanuelNiyonshuti)

• Improved anyio.Path to preserve subclass types by returning Self in methods that return path objects (#1130; PR by @​EmmanuelNiyonshuti)

• Changed the parameter type annotation in anyio.Path.write_bytes() to accept any ReadableBuffer, thus allowing it to accept bytearray and memoryview to match pathlib.Path.write_bytes() (#1135; PR by @​SAY-5)

• Changed several type annotations to only accept callables returning coroutine-like objects instead of arbitrary awaitables:

  • TaskGroup.start_soon()
  • TaskGroup.start()
  • anyio.from_thread.run()

  This reverts an earlier change from v3.7.0 which was made in error. (#1153)

• Changed anyio.run to support callables returning arbitrary awaitables at runtime on all backends. Previously, this only worked on asyncio (#1171; PR by @​gschaffner)

• Changed several classes (and their subclasses) to have __slots__ (with __weakref__):

  • anyio.CancelScope

... (truncated)

Commits

• c384f99 Bumped up the version
• dbba29d Fixed 100% CPU spin on cancel scope misuse (#1217)
• 6bbc6c3 Fix CapacityLimiter over-granting tokens on asyncio (#1172)
• 6f82b25 Refactored TestTLSStream.test_receive_invalid_max_bytes() to be less flaky
• be24b04 Relaxed timeouts to fix test flakiness
• 8113506 Fix test flakiness caused by slow callback duration logging
• 1e988b6 Fixed CapacityLimiter raising trio.WouldBlock instead of anyio.WouldBlock (#1...
• 44713f3 Pin setup-uv to a commit sha across downstream jobs (#1213)
• f1b7301 Fixed stderr writes in a worker subprocess causing a deadlock (#1207)
• 212be93 Fix flaky test_tcp_listener_same_port using a hardcoded port (#1206)
• Additional commits viewable in compare view

Updates urllib3 from 2.7.0 to 2.8.0

Release notes

Sourced from urllib3's releases.

2.8.0

🚀 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. 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

Fixed the following security issues:

• The TLS configuration for HTTPS proxies could be ignored or overridden. (High severity, GHSA-8988-9cw3-xx77)
• HTTPResponse.stream() and read_chunked() could buffer a chunk-size line of unbounded length in memory. (High severity, GHSA-vxq7-64xx-v4gw)
• Chunked Deflate streaming could enter an infinite loop. (Medium severity, GHSA-gh4c-6fx4-qh6g)

[!IMPORTANT] urllib3 2.8.0 fixes HTTPS proxy TLS configuration being ignored or overridden by destination settings. Configurations relying on that behavior may require changes.

Configure proxy CA certificates and client certificates in proxy_ssl_context, and proxy identity checks with proxy_assert_hostname or proxy_assert_fingerprint. Destination client certificates and identity overrides no longer apply to HTTPS forwarding proxy connections.

[!NOTE] CVE IDs had not yet been assigned to these advisories at the time of release due to a backlog at GitHub's CNA.

Deprecations & Removals

• Deprecated using an empty collection as the Retry option allowed_methods to retry any verb. (#5044)

Features

• Added Url.auth_decoded and Url.auth_decoded_joined convenience properties to the result of parse_url(). (#4945)
• Added basic_auth_encoding and proxy_basic_auth_encoding parameters to urllib3.util.make_headers(). (#5092)

Bugfixes

• Fixed response header handling to replace obsolete folded header lines (obs-fold) with spaces in accordance with RFC 9112, preventing raw CRLF sequences from appearing in header values such as Set-Cookie. (#1362)

• Fixed usage of proxy_ssl_context with ProxyManager when use_forwarding_for_https=True. Passing ssl_context instead of proxy_ssl_context for HTTPS proxies in this configuration now emits a FutureWarning and will raise an error in v3.0. (#2577)

• Changed behavior of the default ConnectionPool.pool initialization. LifoQueue is now resolved from the queue module after the ConnectionPool is instantiated instead of using the default cached QueueCls class property. This is done because sometimes the queue.LifoQueue is monkey-patched late in the program, such as by gevent. (#3289)

• Raised UnrewindableBodyError instead of ValueError when retrying a request whose body had tell() but not seek(). (#3779)

• Decoded percent-encoded SOCKS proxy credentials before authenticating with the proxy server. (#3785)

• Fixed HTTPResponse.drain_conn() to discard unread response data in 64 KiB chunks (same as the default amt when doing HTTPResponse.stream(...)). (#5019)

• Fixed is_ipaddress() to detect non-standard IPv4 forms accepted by socket.connect, such as hex (0x7f000001), octal (0177.0.0.1), and decimal integers (2130706433), ensuring SSL certificate verification uses the correct mode for these addresses. (#5029)

• Fixed HTTPConnectionPool.urlopen raising a misleading FullPoolError instead of ValueError when called with an invalid timeout argument on a pool created with block=True. (#5059)

• Fixed port-zero handling to preserve explicit :0 values instead of substituting the default ports 80 or 443 in URL parsing, pool selection, proxy configuration, connection_from_url(), and HTTP/2 request authority. (#5071, #5101)

• Fixed a bug where PoolManager passed the assert_hostname and assert_fingerprint parameters to HTTP connection pools. (#5077)

• Fixed HTTPConnectionPool.urlopen() and HTTP proxy forwarding to strip URL fragments from absolute request targets before sending requests. (#5079)

• Added safeguards to the proxy tunneling code to prevent potential security issues when handling invalid characters in the proxy host and HTTP headers. This change affects users of Python 3.10, Python 3.11, and Python 3.12 when the standard library does not contain the fix; those on newer Python versions should upgrade to 3.13.14+ or 3.14.5+ to get the same security fixes. (#5091)

• Fixed HTTPSConnection.connect() overriding ProxyConfig.ssl_context's certificate policy and proxy identity checks with the target connection's TLS settings when forwarding through an HTTPS proxy.

  HTTPSConnection no longer applies target SNI, assertions, or client credentials to forwarding proxy handshakes and continues to use its ssl_context as a fallback when an HTTPS proxy forwards an HTTP target. (#5093)

• Fixed URL parsing to more strictly enforce RFC 3986 host syntax, rejecting invalid host input such as raw spaces and control characters, malformed percent-encodings, and percent-encoded control characters in HTTP(S) hosts and IPv6 zone identifiers, including proxy CONNECT tunnel targets. Host normalization now also follows RFC 3986 normalization rules for percent-encoded octets by decoding percent-encoded unreserved characters and uppercasing the hexadecimal digits of retained percent-encoded octets. (#5095)

... (truncated)

Changelog

Sourced from urllib3's changelog.

2.8.0 (2026-09-15)

Security

Fixed the following security issues:

• The TLS configuration for HTTPS proxies could be ignored or overridden. (High severity, GHSA-8988-9cw3-xx77 <https://github.com/urllib3/urllib3/security/advisories/GHSA-8988-9cw3-xx77>__)
• HTTPResponse.stream() and read_chunked() could buffer a chunk-size line of unbounded length in memory. (High severity, GHSA-vxq7-64xx-v4gw <https://github.com/urllib3/urllib3/security/advisories/GHSA-vxq7-64xx-v4gw>__)
• Chunked Deflate streaming could enter an infinite loop. (Medium severity, GHSA-gh4c-6fx4-qh6g <https://github.com/urllib3/urllib3/security/advisories/GHSA-gh4c-6fx4-qh6g>__)

.. caution::

urllib3 2.8.0 fixes HTTPS proxy TLS configuration being ignored or
overridden by destination settings. Configurations relying on that
behavior may require changes.
Configure proxy CA certificates and client certificates in
proxy_ssl_context, and proxy identity checks with
proxy_assert_hostname or proxy_assert_fingerprint.
Destination client certificates and identity overrides no longer
apply to HTTPS forwarding proxy connections.

Deprecations & Removals

• Deprecated using an empty collection as the Retry option allowed_methods to retry any verb. ([#5044](https://github.com/urllib3/urllib3/issues/5044) <https://github.com/urllib3/urllib3/issues/5044>__)

Features

• Added Url.auth_decoded and Url.auth_decoded_joined convenience properties to the result of parse_url(). ([#4945](https://github.com/urllib3/urllib3/issues/4945) <https://github.com/urllib3/urllib3/issues/4945>__)
• Added basic_auth_encoding and proxy_basic_auth_encoding parameters to urllib3.util.make_headers(). ([#5092](https://github.com/urllib3/urllib3/issues/5092) <https://github.com/urllib3/urllib3/issues/5092>__)

Bugfixes

... (truncated)

Commits

• b1d30ab Release 2.8.0
• 9016d7e Skip test_read_chunked_with_trailing_data_does_not_hang for brotlicffi (#5258)
• 9101f58 Fix nox -s docs warning (#5256)
• cd770b0 Merge commit from fork
• ea2ad7b Merge commit from fork
• 0716e31 Fix loading unencrypted client keys with a password in pyOpenSSL (#5255)
• 43c68c8 Test pickling of InvalidChunkLength (#5247)
• 308b279 Share security policy between GitHub and Read the Docs (#5253)
• 53fa073 Add policy on duplicate pull requests (#5252)
• 5f2a6a8 Assert on the ALPN extension in test_tunnel_sets_http_11_alpn (#5232)
• 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 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.