Releases: Portkey-AI/gateway
Releases · Portkey-AI/gateway
v1.8.0
What's Changed
- feat: add controlled generations support for google chat by @VisargD in #649
- fix: use token count instead of character count in vertex embeddings … by @VisargD in #657
- feat: lambda provider by @b4s36t4 in #651
- Update Langchain Chatbot with PII Guardrails.ipynb by @siddharthsambharia-portkey in #660
- Adding metadata to webhook guardrails by @roh26it in #655
- fix streaming transformer for perplexity by @narengogi in #654
- Fix:max completion tokens by @shivam-pareek in #667
- add Lemonfoxai by @keshavkrishna in #628
- Add Vercel Cookbook by @siddharthsambharia-portkey in #669
- Feat: portley llm guardrails update by @VisargD in #677
- lowercase-plugin by @thev19 in #648
- feat: add dashscope(alibabacloud) as new provider to gateway by @heimanba in #663
- fix: update logic to extract provider in vertex ai by @narengogi in #682
- feat: allow pydantic support for vertex controlled output generation by @b4s36t4 in #681
- feat: add image generation for google vertex provider by @b4s36t4 in #689
- controlled output - Google by @b4s36t4 in #698
- Feat: Stable diffusion v3 by @narengogi in #615
- Docs: Typo Fix by @Chaitanya110703 in #697
- Hotfix: make transformToFormData optional by @narengogi in #699
- Feature: Bedrock converse integration by @narengogi in #678
- fix response transform for ollama by @narengogi in #710
- Feat: aws assume role setup by @VisargD in #711
- chore: accept anthropic beta and version in body to support environme… by @VisargD in #713
- chore: update aws bedrock assume role logic to allow 2 step process f… by @VisargD in #714
- 1.8.0 by @VisargD in #715
New Contributors
- @thev19 made their first contribution in #648
- @heimanba made their first contribution in #663
- @Chaitanya110703 made their first contribution in #697
Full Changelog: v1.7.7...v1.8.0
v1.7.7
What's Changed
- chore: upgrade portkey mustache package version to 2.1.2 by @VisargD in #611
- Chore/support additional params cohere embed by @narengogi in #614
- Add missing logprob mapping for Azure OpenAI provider by @rootAvish in #618
- Added Support for the new /fim/completions route on Mistral AI API by @keshavkrishna in #467
- fix: chat templating for llama and mistral hosted on bedrock by @narengogi in #619
- Add contextual embedding cookbook by @siddharthsambharia-portkey in #627
- webhook plugin header parsing logic to allow objects by @VisargD in #632
- Ends With Plugin by @shivam-pareek in #607
- fix: added max_completion_tokens for chat complete by @shivam-pareek in #617
- Chore: Debug configs for vscode by @narengogi in #602
- fix: send Portkey as user-agent by @b4s36t4 in #609
- Add default models for gemini, together ai, fireworks ai by @vrushankportkey in #630
- Upstage.ai by @shivam-pareek in #608
- fix chat templating for llama hosted on bedrock by @narengogi in #623
- improvement: Add vertex embeddings support by @elentaure in #622
- Fix media file prompting for vertex gemini and google gemini by @narengogi in #635
- feat: add json mode support for vertex-ai by @VisargD in #638
- Chore: better error handling by @VisargD in #640
- Feat: bedrock cross region model ids by @VisargD in #641
- chore: update plugin manifest with ends with handler by @VisargD in #645
- 1.7.7 by @VisargD in #646
New Contributors
- @rootAvish made their first contribution in #618
- @keshavkrishna made their first contribution in #467
- @shivam-pareek made their first contribution in #607
- @elentaure made their first contribution in #622
Full Changelog: v1.7.6...v1.7.7
v1.7.6
v1.7.5
What's Changed
- fix: move sambanova fastapi to cloud by @ilsubyeega in #592
- Feat all uppercase plugin by @shaun-ak in #570
- feat: add inference-net as a provider by @b4s36t4 in #587
- Minor fixes by @narengogi in #595
- chore: update portkey mustache package version to 2.1.0 by @VisargD in #598
- 1.7.5 by @VisargD in #600
New Contributors
Full Changelog: v1.7.4...v1.7.5
v1.7.4
What's Changed
- added new cookbooks by @satvik314 in #469
- Guardrails cookbook by @siddharthsambharia-portkey in #573
- Guardrails cookbook by @siddharthsambharia-portkey in #577
- add nudge to star the repo by @vrushankportkey in #578
- fix: type support for default values by @b4s36t4 in #583
- provider: SambaNova FastAPI implementation. by @ilsubyeega in #548
- fix: change message content format for aporia guardrail by @b4s36t4 in #581
- Enhance JSON Schema and JSON Keys plugins with improved error handling and explanations by @roh26it in #586
- feat: combine system messages for anthropic by @jpulec in #575
- fix: aporia guardrails request messages mapping by @VisargD in #589
- 1.7.4 by @VisargD in #591
Full Changelog: v1.7.3...v1.7.4
v1.7.3
What's Changed
- Fix cohere message param transform by @narengogi in #562
- Fix/monster chat completions by @narengogi in #564
- fix: wrong provider & lint by @b4s36t4 in #566
- refactor code for minimal changes required for new open-ai compatible… by @b4s36t4 in #565
- fix: update gemini image_url parsing logic to support other mime type… by @VisargD in #567
- 1.7.3 by @VisargD in #569
Full Changelog: v1.7.2...v1.7.3
v1.7.2
What's Changed
- update cohere endpoints & response template by @b4s36t4 in #415
- Feat/azure inference support by @b4s36t4 in #534
- Maintain a mapping for file extension and mimetype for vertex by @narengogi in #552
- fix: aporia request messages context reference by @VisargD in #555
- chore: support api endpoint configuration for aporia plugin to allow … by @VisargD in #553
- chore: update patronus response schema based on its field deprecation… by @VisargD in #554
- 1.7.2 by @VisargD in #559
- fix: skip husky prepare script for production environment by @VisargD in #560
Full Changelog: v1.7.1...v1.7.2
v1.7.1
What's Changed
- Add provider config and endpoints for azure audio routes by @narengogi in #544
- Import middleware signed by @MuraraAllan in #526
- docs: fix linking typo on README(.cn).md by @ilsubyeega in #547
- Add support for Huggingface's openAI like endpoint by @narengogi in #524
- feat: allow to accept beta for anthropic, support for prompt caching by @b4s36t4 in #519
- [Provider]: Deepbricks, SiliconFlow by @sternelee in #533
- Feat/triton support by @VisargD in #549
- 1.7.1 by @VisargD in #550
New Contributors
- @MuraraAllan made their first contribution in #526
- @ilsubyeega made their first contribution in #547
Full Changelog: v1.7.0...v1.7.1
v1.7.0
What's Changed
- Patronus checks added by @roh26it in #503
- Create Contributing.md by @siddharthsambharia-portkey in #508
- Added a logging check by @roh26it in #517
- chore: allow async pre request validator by @VisargD in #522
- Voyage api support by @b4s36t4 in #506
- Voyage Integration by @Miraj98 in #399
- Deepseek support by @b4s36t4 in #505
- Provider: add deepseek support by @nsdou in #359
- Add minimal linting by @narengogi in #520
- Add X-Title header for openrouter requests by @narengogi in #528
- fix: pass response transformer type for cached stream response handler by @VisargD in #527
- Fix racias spelling - update manifest.json by @vrushankportkey in #530
- fix: Make anthropic usage object in streaming mode compliant with openAI response by @narengogi in #497
- 1.6.1 by @VisargD in #532
- Update README.md with LangChain Agent Integration by @siddharthsambharia-portkey in #536
- Support Vertex Llama Hosted API Enpoint by @narengogi in #484
- Feat: Provider Tests by @narengogi in #494
- Remove a few unused variables by @narengogi in #521
- Feat: Conditional Router by @VisargD in #539
- Azure OpenAI Audio routes by @narengogi in #538
- fix: support request content array in guardrails extract request text… by @VisargD in #542
- 1.7.0 by @VisargD in #543
New Contributors
- @b4s36t4 made their first contribution in #506
- @Miraj98 made their first contribution in #399
- @nsdou made their first contribution in #359
Full Changelog: v1.6.0...v1.7.0