v1.2.12

What's Changed

• Upgrade openai-fetch and OpenAI pricing by @rileytomasek in #30
• Upgrade dependencies by @rileytomasek in #31

Full Changelog: v1.2.11...v1.2.12

Remove extra console.warn for non-openai models

• fix: temporarily link 'np' locally bf29067
• feat: update np to latest d575c67
• fix: remove console.warn about missing cost (#29) 71aadb2

v1.2.10...v1.2.11

v1.2.10

What's Changed

• Ensure embeddings are same size as input by @stevepeak in #27

New Contributors

• @stevepeak made their first contribution in #27

Full Changelog: v1.2.9...v1.2.10

v1.2.9

What's Changed

• feat: Allow async cacheKey option by @colelawrence in #26

New Contributors

• @colelawrence made their first contribution in #26

Full Changelog: v1.2.8...v1.2.9

v1.2.8

• Update error for tool_calls validation errors 1058b8e

v1.2.7...v1.2.8

v1.2.7

What's Changed

• Improve error handling by @rileytomasek in #23

Full Changelog: v1.2.6...v1.2.7

v1.2.6

• Add latency to Emedding model response cc8f333

Full Changelog: v1.2.5...v1.2.6

v1.2.5

• Add missing latency to SparseVector model response 58c42c4

Full Changelog: v1.2.4...v1.2.5

v1.2.4

What's Changed

• Standardize package.json scripts by @rileytomasek in #15
• Remove duplicate params & context args by @rileytomasek in #14
• Minor fixes to GPT tokenizer by @rileytomasek in #16
• Fix datastore to rely on same path to embeddings by @rileytomasek in #19

Full Changelog: v1.2.3...v1.2.4

change defaultCacheKey to use sha512 instead of sha256

v1.2.3

1.2.3