feat(minimax): add image/audio/video/music model entries (#77)

Extend modality enum to support video and music, then register the
non-text MiniMax models that were already declared in
media_capabilities but had no concrete entries:

- image-01 ($0.0035/image)
- speech-2.8/2.6 hd & turbo ($60-$100 per 1M chars)
- Hailuo 2.3 Fast / 2.3 / 02 video models ($0.10-$0.56 per video)
- music-2.6, lyrics_generation

Per-call pricing is documented in inline comments since the schema's
token-based cost fields don't naturally fit per-call billing.

schema.toml and scripts/validate.py both updated; the change is
additive (existing modality values remain valid).

providers/minimax.toml

@@ -1,6 +1,11 @@
 # MiniMax — https://minimax.io
-# Models: 4
+# Models: 14 (4 text + 1 image + 4 audio TTS + 3 video + 2 music)
 # Regions: international (default), china (minimaxi.com)
+#
+# Non-text models are billed per call rather than per token, so
+# input_cost_per_m / output_cost_per_m are set to 0.0 and the actual
+# rate is documented in the inline comment above each entry.
+# Source: https://platform.minimax.io/docs/guides/pricing-paygo
 
 [provider]
 id = "minimax"
@@ -70,3 +75,141 @@ supports_tools = true
 supports_vision = true
 supports_streaming = true
 aliases = ["minimax-m2.5-highspeed", "m2.5-highspeed"]
+
+# --- Image generation ---
+
+# $0.0035 per image
+[[models]]
+id = "image-01"
+display_name = "MiniMax Image 01"
+tier = "smart"
+modality = "image"
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = false
+aliases = ["minimax-image"]
+
+# --- Text-to-speech ---
+
+# $100 per 1M characters
+[[models]]
+id = "speech-2.8-hd"
+display_name = "MiniMax Speech 2.8 HD"
+tier = "frontier"
+modality = "audio"
+input_cost_per_m = 100.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = true
+aliases = ["minimax-tts-hd", "speech-hd"]
+
+# $60 per 1M characters
+[[models]]
+id = "speech-2.8-turbo"
+display_name = "MiniMax Speech 2.8 Turbo"
+tier = "fast"
+modality = "audio"
+input_cost_per_m = 60.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = true
+aliases = ["minimax-tts-turbo", "speech-turbo"]
+
+# $100 per 1M characters
+[[models]]
+id = "speech-2.6-hd"
+display_name = "MiniMax Speech 2.6 HD"
+tier = "smart"
+modality = "audio"
+input_cost_per_m = 100.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = true
+aliases = []
+
+# $60 per 1M characters
+[[models]]
+id = "speech-2.6-turbo"
+display_name = "MiniMax Speech 2.6 Turbo"
+tier = "fast"
+modality = "audio"
+input_cost_per_m = 60.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = true
+aliases = []
+
+# --- Video generation (Hailuo) ---
+
+# Per-call: $0.19 (768P/6s) up to $0.33 (1080P/6s)
+[[models]]
+id = "MiniMax-Hailuo-2.3-Fast"
+display_name = "Hailuo 2.3 Fast"
+tier = "fast"
+modality = "video"
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = false
+aliases = ["hailuo-2.3-fast", "hailuo-fast"]
+
+# Per-call: $0.28 (768P/6s) up to $0.56 (1080P/6s)
+[[models]]
+id = "MiniMax-Hailuo-2.3"
+display_name = "Hailuo 2.3"
+tier = "frontier"
+modality = "video"
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = false
+aliases = ["hailuo-2.3", "hailuo"]
+
+# Per-call: $0.10 (512P/6s) up to $0.56 (1080P/6s)
+[[models]]
+id = "MiniMax-Hailuo-02"
+display_name = "Hailuo 02"
+tier = "smart"
+modality = "video"
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = false
+aliases = ["hailuo-02"]
+
+# --- Music generation ---
+
+# $0.15 per up-to-5-minute track
+[[models]]
+id = "music-2.6"
+display_name = "MiniMax Music 2.6"
+tier = "frontier"
+modality = "music"
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = false
+aliases = ["minimax-music"]
+
+# $0.01 per song
+[[models]]
+id = "lyrics_generation"
+display_name = "MiniMax Lyrics Generation"
+tier = "fast"
+modality = "music"
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_tools = false
+supports_vision = false
+supports_streaming = false
+aliases = ["minimax-lyrics"]

schema.toml

@@ -75,8 +75,8 @@ example = "smart"
 [provider.sections.models.fields.modality]
 type = "string"
 required = false
-description = "Model modality. 'text' (default) is a chat/LLM; 'image' is an image-generation model (context_window/max_output_tokens become optional); 'audio' is a speech model."
+description = "Model modality. 'text' (default) is a chat/LLM; 'image' is an image-generation model; 'audio' is a speech / TTS model; 'video' is a video-generation model; 'music' is a music / lyrics generation model. For non-text modalities, context_window / max_output_tokens are optional and input_cost_per_m / output_cost_per_m may be 0 when the model is billed per call."
-options = ["text", "image", "audio"]
+options = ["text", "image", "audio", "video", "music"]
 default = "text"
 example = "text"
 

scripts/validate.py

@@ -34,7 +34,7 @@ except ImportError:
         sys.exit(1)
 
 VALID_TIERS = {"frontier", "smart", "balanced", "fast", "local"}
-VALID_MODALITIES = {"text", "image", "audio"}
+VALID_MODALITIES = {"text", "image", "audio", "video", "music"}
 VALID_HAND_CATEGORIES = {
     "communication", "content", "data", "development",
     "devops", "finance", "productivity", "research", "social",