What happened?
I get "prompt enhancement failed: name 're' is not defined" when trying to enhance the prompt in H3 Omni, only in Omni.
Your system (OS / GPU / RAM / VRAM / etc.)
Windows 11 32GB DDR4 RTX 5060TI 16GB
Logs / full error output
[Enhance] LoRA check: activated_loras=[], model_type=minimax_h3_ref2va_fused_turbo
[Enhance] Loaded guide: minimax_h3_ref2va_video.md (has_images=True)
[Enhance] Using model-specific guide for minimax_h3_ref2va_fused_turbo (video, images=True)
[Enhance] Sending 1 image(s) to vision LLM
[LLM] Resized image for LLM: 3840x2160 -> 768x432
[LLM] Response: 682 tokens generated (prompt=2527, finish=stop)
[LLM] Completion telemetry: reasoning=0 tokens, answer=682 tokens, effort=off/default, budget=0, finish=stop
INFO: 127.0.0.1:12745 - "POST /api/v1/llm/enhance-prompt HTTP/1.1" 500 Internal Server Error
INFO: 127.0.0.1:2492 - "GET /api/v1/loras/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:2492 - "PUT /api/v1/studio-preferences HTTP/1.1" 200 OK
INFO: 127.0.0.1:6481 - "GET /api/v1/model-options/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "GET /api/v1/defaults/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:6481 - "GET /api/v1/loras/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "GET /api/v1/loras/minimax_h3_ref2va/details HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "GET /api/v1/loras/minimax_h3_ref2va/details HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "PUT /api/v1/studio-preferences HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "PUT /api/v1/studio-preferences HTTP/1.1" 200 OK
[Enhance] LoRA check: activated_loras=['MiniMax-H3-Ref2VA-Acc-8Step.safetensors'], model_type=minimax_h3_ref2va
[Enhance] LoRA dir: loras\minimax_h3
[Enhance] LoRA 'MiniMax-H3-Ref2VA-Acc-8Step.safetensors': triggers=[], sidecar=False
[Enhance] No LoRA triggers extractable from 1 LoRA(s)
[Enhance] Loaded guide: minimax_h3_ref2va_video.md (has_images=True)
[Enhance] Using model-specific guide for minimax_h3_ref2va (video, images=True)
[Enhance] Sending 1 image(s) to vision LLM
[LLM] Resized image for LLM: 11520x6528 -> 768x435
[LLM] Response: 630 tokens generated (prompt=2440, finish=stop)
[LLM] Completion telemetry: reasoning=0 tokens, answer=630 tokens, effort=off/default, budget=0, finish=stop
INFO: 127.0.0.1:4661 - "POST /api/v1/llm/enhance-prompt HTTP/1.1" 500 Internal Server Error
[Enhance] LoRA check: activated_loras=['MiniMax-H3-Ref2VA-Acc-8Step.safetensors'], model_type=minimax_h3_ref2va
[Enhance] LoRA dir: loras\minimax_h3
[Enhance] LoRA 'MiniMax-H3-Ref2VA-Acc-8Step.safetensors': triggers=[], sidecar=False
[Enhance] No LoRA triggers extractable from 1 LoRA(s)
[Enhance] Loaded guide: minimax_h3_ref2va_video.md (has_images=True)
[Enhance] Using model-specific guide for minimax_h3_ref2va (video, images=True)
[Enhance] Sending 1 image(s) to vision LLM
[LLM] Resized image for LLM: 11520x6528 -> 768x435
[LLM] Response: 624 tokens generated (prompt=2440, finish=stop)
[LLM] Completion telemetry: reasoning=0 tokens, answer=624 tokens, effort=off/default, budget=0, finish=stop
INFO: 127.0.0.1:4633 - "POST /api/v1/llm/enhance-prompt HTTP/1.1" 500 Internal Server Error
What happened?
I get "prompt enhancement failed: name 're' is not defined" when trying to enhance the prompt in H3 Omni, only in Omni.
Your system (OS / GPU / RAM / VRAM / etc.)
Windows 11 32GB DDR4 RTX 5060TI 16GB
Logs / full error output
[Enhance] LoRA check: activated_loras=[], model_type=minimax_h3_ref2va_fused_turbo
[Enhance] Loaded guide: minimax_h3_ref2va_video.md (has_images=True)
[Enhance] Using model-specific guide for minimax_h3_ref2va_fused_turbo (video, images=True)
[Enhance] Sending 1 image(s) to vision LLM
[LLM] Resized image for LLM: 3840x2160 -> 768x432
[LLM] Response: 682 tokens generated (prompt=2527, finish=stop)
[LLM] Completion telemetry: reasoning=0 tokens, answer=682 tokens, effort=off/default, budget=0, finish=stop
INFO: 127.0.0.1:12745 - "POST /api/v1/llm/enhance-prompt HTTP/1.1" 500 Internal Server Error
INFO: 127.0.0.1:2492 - "GET /api/v1/loras/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:2492 - "PUT /api/v1/studio-preferences HTTP/1.1" 200 OK
INFO: 127.0.0.1:6481 - "GET /api/v1/model-options/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "GET /api/v1/defaults/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:6481 - "GET /api/v1/loras/minimax_h3_ref2va HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "GET /api/v1/loras/minimax_h3_ref2va/details HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "GET /api/v1/loras/minimax_h3_ref2va/details HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "PUT /api/v1/studio-preferences HTTP/1.1" 200 OK
INFO: 127.0.0.1:4661 - "PUT /api/v1/studio-preferences HTTP/1.1" 200 OK
[Enhance] LoRA check: activated_loras=['MiniMax-H3-Ref2VA-Acc-8Step.safetensors'], model_type=minimax_h3_ref2va
[Enhance] LoRA dir: loras\minimax_h3
[Enhance] LoRA 'MiniMax-H3-Ref2VA-Acc-8Step.safetensors': triggers=[], sidecar=False
[Enhance] No LoRA triggers extractable from 1 LoRA(s)
[Enhance] Loaded guide: minimax_h3_ref2va_video.md (has_images=True)
[Enhance] Using model-specific guide for minimax_h3_ref2va (video, images=True)
[Enhance] Sending 1 image(s) to vision LLM
[LLM] Resized image for LLM: 11520x6528 -> 768x435
[LLM] Response: 630 tokens generated (prompt=2440, finish=stop)
[LLM] Completion telemetry: reasoning=0 tokens, answer=630 tokens, effort=off/default, budget=0, finish=stop
INFO: 127.0.0.1:4661 - "POST /api/v1/llm/enhance-prompt HTTP/1.1" 500 Internal Server Error
[Enhance] LoRA check: activated_loras=['MiniMax-H3-Ref2VA-Acc-8Step.safetensors'], model_type=minimax_h3_ref2va
[Enhance] LoRA dir: loras\minimax_h3
[Enhance] LoRA 'MiniMax-H3-Ref2VA-Acc-8Step.safetensors': triggers=[], sidecar=False
[Enhance] No LoRA triggers extractable from 1 LoRA(s)
[Enhance] Loaded guide: minimax_h3_ref2va_video.md (has_images=True)
[Enhance] Using model-specific guide for minimax_h3_ref2va (video, images=True)
[Enhance] Sending 1 image(s) to vision LLM
[LLM] Resized image for LLM: 11520x6528 -> 768x435
[LLM] Response: 624 tokens generated (prompt=2440, finish=stop)
[LLM] Completion telemetry: reasoning=0 tokens, answer=624 tokens, effort=off/default, budget=0, finish=stop
INFO: 127.0.0.1:4633 - "POST /api/v1/llm/enhance-prompt HTTP/1.1" 500 Internal Server Error