libs/ai-endpoints/v0.1.6

Full Changelog: libs/ai-endpoints/v0.1.5...libs/ai-endpoints/v0.1.6

libs/ai-endpoints/v0.1.5

What's Changed

• update default chat model to meta/llama3-8b-instruct (most features & deployment options) by @mattf in #73
• add support for reranking api change w/ truncate parameter by @mattf in #75

Full Changelog: libs/ai-endpoints/v0.1.4...libs/ai-endpoints/v0.1.5

libs/ai-endpoints/v0.1.4

What's Changed

• skip nvolveqa_40k compat test when run against local nim by @mattf in #70
• Scheduled test cases fix by @raspawar in #71
• update ranking nim support for nims w/ /v1/models and multiple names by @mattf in #69

Full Changelog: libs/ai-endpoints/v0.1.3...libs/ai-endpoints/v0.1.4

libs/ai-endpoints/v0.1.3

What's Changed

• make stop property functional by @mattf in #52
• add scheduled tests by @ccurme in #53
• Name scheduled test run by @efriis in #54
• nim docs model name update by @dglogo in #55
• no models support steerlm params, removing references by @mattf in #56
• add rerank example using duckduckgo, bm25 and faiss by @mattf in #47
• Fix for failing lint workflow by @raspawar in #58
• Provide default model in local NIM mode by @raspawar in #51
• Fix bug in Embedding/Rerank default model by @raspawar in #60
• refactor payload preprocessing by @mattf in #61
• refactor payload preprocessing (continued) by @mattf in #62
• ai-endpoints[patch]: add mypy overrides by @ccurme in #64
• add register_model for users to add their own custom models with endpoints [api change; expansion] by @mattf in #57
• release permissions by @efriis in #66
• remove broken stop word detection, rely on server side implementation by @mattf in #63

New Contributors

• @ccurme made their first contribution in #53

Full Changelog: libs/ai-endpoints/v0.1.2...libs/ai-endpoints/v0.1.3

libs/ai-endpoints/v0.1.2

What's Changed

• Fix for NVIDIAClient output to redact the api_key by @raspawar in #50

New Contributors

• @raspawar made their first contribution in #50

Full Changelog: libs/ai-endpoints/v0.1.1...libs/ai-endpoints/v0.1.2

libs/ai-endpoints/v0.1.1

What's Changed

• make connectors use user provided base_url when the provided model is also hosted by @mattf in #49

Full Changelog: libs/ai-endpoints/v0.1.0...libs/ai-endpoints/v0.1.1

libs/ai-endpoints/v0.1.0

What's Changed

• update to 0.1, remove deprecated functionality and focus on api catalog backend by @mattf in #48

Full Changelog: libs/ai-endpoints/v0.0.20...libs/ai-endpoints/v0.1.0

this is version 0.1.0 of the connectors with two primary changes -

1. [user visible] deprecated or unavailable functionality removed
2. [not use visible] use of api catalog (integrate.api.nvidia.com and ai.api.nvidia.com) instead of nvcf (api.nvcf.nvidia.com) for inference

all playground_* model endpoints have been decommissioned.

functionality removed -

• models: playground_mamba_chat, playground_smaug_72b, playground_nemotron_qa_8b, playground_nemotron_steerlm_8b, playground_steerlm_llama_70b, playground_yi_34b, playground_nvolveqa_40k
• methods: available_functions, get_available_functions, get_model_details, get_binding_model, mode, validate_model, validate_base_url, reset_method_cache, validate_client, aifm_deprecated, aifm_bad_deprecated, aifm_labels_deprecated, custom_preprocess, preprocess_msg, custom_postprocess, get_generation, get_stream, get_astream, get_payload, prep_payload, prep_msg, deprecate_max_length
• properties: ChatNVIDIA.bad, ChatNVIDIA.labels, ChatNVIDIA.infer_endpoint, ChatNVIDIA.client, ChatNVIDIA.streaming, NVIDIAEmbeddings.infer_endpoint, NVIDIAEmbeddings.client, NVIDIAEmbeddings.max_length

functionality deprecated -

• NVIDIAEmbeddings.model_type, instead of setting model_type="query" or "passage" use NVIDIAEmbeddings.embed_query and NVIDIAEmbeddings.embed_documents

migration guide -

• ChatNVIDIA().mode("nim", base_url="http://...") must use ChatNVIDIA(base_url="http://...")
• NVIDIAEmbeddings().mode("nim", base_url="http://...") must use NVIDIAEmbeddings(base_url="http://...")
• NVIDIARerank().mode("nim", base_url="http://...") must use NVIDIARerank(base_url="http://...")
• compatibility for the playground_nvolveqa_40k (aka nvolveqa_40k) model. when specifying model="nvolveqa_40k", the NV-Embed-QA model will be used and truncate="END" will be set. note: the old nvolveqa_40k model endpoint would silently truncated input while all available endpoints raise an error instead of truncating.

model migration, the following models will raise a warning and use an alternative (model changes in italics) -

model	alternative
playground_llama2_13b	meta/llama2-70b
playground_llama2_code_13b	meta/codellama-70b
playground_llama2_code_34b	meta/codellama-70b
playground_nv_llama2_rlhf_70b	meta/llama2-70b
playground_phi2	microsoft/phi-3-mini-4k-instruct
playground_llama2_code_70b	meta/codellama-70b
playground_gemma_2b	google/gemma-2b
playground_gemma_7b	google/gemma-7b
playground_llama2_70b	meta/llama2-70b
playground_mistral_7b	mistralai/mistral-7b-instruct-v0.2
playground_mixtral_8x7b	mistralai/mixtral-8x7b-instruct-v0.1
playground_deplot	google/deplot
playground_fuyu_8b	adept/fuyu-8b
playground_kosmos_2	microsoft/kosmos-2
playground_neva_22b	nvidia/neva-22b
playground_nvolveqa_40k	NV-Embed-QA

libs/ai-endpoints/v0.0.20

What's Changed

• fix pypi links by @efriis in #45
• Dglog/docs updates 2 by @dglogo in #46
• cookbook: add langgraph nim demo by @efriis in #42

Full Changelog: libs/ai-endpoints/v0.0.19...libs/ai-endpoints/v0.0.20

libs/ai-endpoints/v0.0.19

Full Changelog: libs/ai-endpoints/v0.0.18...libs/ai-endpoints/v0.0.19

libs/ai-endpoints/v0.0.18

What's Changed

• Deprecate mode() in favor of init(base_url=...) by @mattf in #43

Full Changelog: libs/ai-endpoints/v0.0.17...libs/ai-endpoints/v0.0.18