# LiteLLM v1.99.0-dev.1

- Product: LiteLLM (https://whatsnew.fyi/product/litellm)
- Vendor: BerriAI
- Date: 2026-08-19
- Version: v1.99.0-dev.1
- Original notes: https://github.com/BerriAI/litellm/releases/tag/v1.99.0-dev.1
- Permalink: https://whatsnew.fyi/product/litellm/releases/v1.99.0-dev.1
- Labels: Pre-release

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **security** — Docker images are signed with cosign for verification
- **fixed** — Register WebSocket passthrough for OpenAI prefixes in proxy
- **fixed** — Report uploaded size in the FileObject returned by Bedrock managed batch uploads
- **fixed** — Support AWS Bedrock batch cancellation via StopModelInvocationJob
- **added** — Async Rust OCR Bridge and MCP OAuth UI Restore
- **fixed** — Do not crash logging when a completed batch has no output file
- **fixed** — Add default model pin to complexity router UI
- **added** — Add Lite mixed-provider auto-router preset in UI
- **added** — Link key info header to its user, creator, team, and organization in UI
- **fixed** — Scan text on /guardrails/apply_guardrail for Azure Content Safety
- **added** — Forward LiteLLM identity and metadata into Bedrock requestMetadata
- **fixed** — Rename max_tokens to max_completion_tokens for gpt-5-chat deployments in Azure
- **fixed** — Preserve cache token usage when invocationMetrics replace the usage block in Bedrock
- **fixed** — Registry caches stop per-request tag and end-user Postgres reads in auth
- **fixed** — Return 400 naming the missing required param on POST /v1/batches
- **security** — Bump sqlparse to 0.6.0 to resolve osv-scan CVEs
- **fixed** — Stop pairing key spend with the team budget when a key has no budget in UI
- **fixed** — Record MCP tool guardrail evaluations and blocks
- **fixed** — Return 400 for non-object metadata and litellm_metadata instead of silent drop or 500 in proxy
- **fixed** — Preserve optional Responses tool properties in Anthropic

##### Verify Docker Image Signature

All LiteLLM Docker images are signed with [cosign](https://docs.sigstore.dev/cosign/overview/). Every release is signed with the same key introduced in [commit `0112e53`](https://github.com/BerriAI/litellm/commit/0112e53046018d726492c814b3644b7d376029d0).

**Verify using the pinned commit hash (recommended):**

A commit hash is cryptographically immutable, so this is the strongest way to ensure you are using the original signing key:

```bash
cosign verify \
  --key https://raw.githubusercontent.com/BerriAI/litellm/0112e53046018d726492c814b3644b7d376029d0/cosign.pub \
  ghcr.io/berriai/litellm:v1.99.0-dev.1
```

**Verify using the release tag (convenience):**

Tags are protected in this repository and resolve to the same key. This option is easier to read but relies on tag protection rules:

```bash
cosign verify \
  --key https://raw.githubusercontent.com/BerriAI/litellm/v1.99.0-dev.1/cosign.pub \
  ghcr.io/berriai/litellm:v1.99.0-dev.1
```

Expected output:

```
The following checks were performed on each of these signatures:
  - The cosign claims were validated
  - The signatures were verified against the specified public key
```

---
##### What's Changed
* chore(typing): drop 1.3k basedpyright errors across 30 Any hotspot files by @mateo-berri in https://github.com/BerriAI/litellm/pull/37073
* fix(proxy): register WebSocket passthrough for OpenAI prefixes by @LHMQ878 in https://github.com/BerriAI/litellm/pull/36151
* fix(bedrock): report uploaded size in the FileObject returned by managed batch uploads by @devin-ai-integration[bot] in https://github.com/BerriAI/litellm/pull/36392
* fix(batches): support AWS Bedrock batch cancellation via `StopModelInvocationJob` by @ArjunPakhan in https://github.com/BerriAI/litellm/pull/34087
* feat: Async Rust OCR Bridge and MCP OAuth UI Restore by @ArjunPakhan in https://github.com/BerriAI/litellm/pull/31453
* fix(batches): don't crash logging when a completed batch has no output file by @MUSE-CODE-SPACE in https://github.com/BerriAI/litellm/pull/34067
* fix(UI): add default model pin to complexity router UI by @tin-berri in https://github.com/BerriAI/litellm/pull/36615
* feat(ui): add Lite mixed-provider auto-router preset by @tin-berri in https://github.com/BerriAI/litellm/pull/37068
* feat(ui): link key info header to its user, creator, team, and organization by @ryan-crabbe-berri in https://github.com/BerriAI/litellm/pull/37187
* fix(guardrails): scan text on /guardrails/apply_guardrail for Azure Content Safety by @yucheng-berri in https://github.com/BerriAI/litellm/pull/36894
* feat(bedrock): forward LiteLLM identity and metadata into Bedrock requestMetadata by @yassin-berriai in https://github.com/BerriAI/litellm/pull/36861
* fix(azure): rename max_tokens to max_completion_tokens for gpt-5-chat deployments by @yassin-berriai in https://github.com/BerriAI/litellm/pull/36857
* fix(bedrock): preserve cache token usage when invocationMetrics replace the usage block by @brian5021 in https://github.com/BerriAI/litellm/pull/36878
* fix(proxy): registry caches stop per-request tag and end-user Postgres reads in auth by @ryan-crabbe-berri in https://github.com/BerriAI/litellm/pull/36801
* test(e2e): replay a real tool-search assistant turn back to Bedrock Invoke by @yassin-berriai in https://github.com/BerriAI/litellm/pull/36856
* fix(proxy): return 400 naming the missing required param on POST /v1/batches by @mateo-berri in https://github.com/BerriAI/litellm/pull/37199
* fix(ci): bump sqlparse to 0.6.0 to resolve osv-scan CVEs by @mateo-berri in https://github.com/BerriAI/litellm/pull/37200
* fix(ui): stop pairing key spend with the team budget when a key has no budget by @ryan-crabbe-berri in https://github.com/BerriAI/litellm/pull/37196
* fix(guardrails): record MCP tool guardrail evaluations and blocks in … by @Scott-Wilson-ZocDoc in https://github.com/BerriAI/litellm/pull/36978
* fix(proxy): return 400 for non-object metadata and litellm_metadata in

_[Truncated at 4000 characters — full notes: https://github.com/BerriAI/litellm/releases/tag/v1.99.0-dev.1]_