v9.0.6 - metadata hotfix

What's Changed

• nsfw info gen_metadata fix by @rbrtcs1 in #297
• fix: mark all non-censored nsfw images in info by @db0 in #300

Full Changelog: v9.0.5...v9.0.6

v9.0.5 - fix censorship reporting; add `nsfw` information

What's Changed

• fix: more accurate gen_metadata logic; always send nsfw information by @rbrtcs1 in #295

New Contributors

• @rbrtcs1 made their first contribution in #295

Full Changelog: v9.0.4...v9.0.5

v9.0.4 - gen_metadata fixes; `horde_url` config option now works

What's Changed

• fix: correct report of censor by @db0 in #293
• fix: set AI_HORDE_URL before horde_sdk import by @tazlin in #291
• WIP: Update documentation for civitai_api_token by @CIB in #282

New Contributors

• @CIB made their first contribution in #282

Full Changelog: v9.0.3...v9.0.4

v9.0.3 - Flux meta model load hotfix

What's Changed

• feat: load_large_models for model meta by @tazlin in #289

Full Changelog: v9.0.2...v9.0.3

v9.0.2 - Flux support

What's Changed

• feat: allow setting worker name via env var or cli arg by @tazlin in #245
• fix: run conda commands relative to linux script by @tazlin in #252
• fix: amd/linux script improvements; exit_on_unhandled_faults improvement by @tazlin in #258
• feat: latest comfyui; fix: better GPU utilization for SD15 by @tazlin in #270

Full Changelog: v8.1.2...v9.0.2

v8.1.2

What's Changed

• fix: asyncio win loop policy fix by @tazlin in #244

Full Changelog: v8.1.1...v8.1.2

v8.1.1

What's Changed

• Add workers' need for SSD to README by @RadixSeven in #227
• Initial AMD GPU Support by @niales in #223
• feat: Adds support for generating transparent images via layer diffusion by @db0 in #224

New Contributors

• @RadixSeven made their first contribution in #227
• @niales made their first contribution in #223

Full Changelog: v7.0.4...v8.1.1

v7.0.4

Full Changelog: v7.0.3...v7.0.4

v7.0.3

What's Changed

• fix: pin numpy to 1.26.4 (v2 is breaking) by @tazlin in #214

Full Changelog: v7.0.2...v7.0.3

v7.0.2

What's Changed

• fix: handle custom models in keep_single_inference by @tazlin in #204

Full Changelog: v7.0.1...v7.0.2