Fix API URL (api.greenpt.ai), add models command, remove debug logging

assistant/assistant.py

@@ -1,11 +1,8 @@
-import logging
 import aiohttp
 import discord
 from redbot.core import commands, Config, checks
 from redbot.core.bot import Red
 
-log = logging.getLogger("red.assistant")
-
 
 class APIKeyModal(discord.ui.Modal):
     def __init__(self, config):
@@ -38,7 +35,7 @@ class APIKeyButton(discord.ui.View):
 class Assistant(commands.Cog):
     """AI chatbot assistant. Set a chat channel or get responses when the bot is pinged."""
 
-    BASE_URL = "https://greenpt.com/v1"
+    BASE_URL = "https://api.greenpt.ai/v1"
 
     def __init__(self, bot: Red):
         self.bot = bot
@@ -75,19 +72,12 @@ class Assistant(commands.Cog):
 
         guild = message.guild
 
-        if self.bot.user in message.mentions:
-            log.info("[ASSISTANT DEBUG] Mention detected from %s in #%s", message.author, message.channel)
-
         enabled = await self.config.guild(guild).enabled()
         if not enabled:
-            if self.bot.user in message.mentions:
-                log.info("[ASSISTANT DEBUG] Guild %s is disabled", guild.id)
             return
 
         api_key = await self.config.api_key()
         if not api_key:
-            if self.bot.user in message.mentions:
-                log.info("[ASSISTANT DEBUG] No API key set")
             return
 
         chat_channel_id = await self.config.guild(guild).chat_channel()
@@ -99,8 +89,6 @@ class Assistant(commands.Cog):
         if not is_chat_channel and not is_ping:
             return
 
-        log.info("[ASSISTANT DEBUG] Processing message, is_ping=%s, is_chat_channel=%s", is_ping, is_chat_channel)
-
         content = message.content
         if is_ping:
             for mention in (f"<@{self.bot.user.id}>", f"<@!{self.bot.user.id}>"):
@@ -225,6 +213,93 @@ class Assistant(commands.Cog):
         await self.config.guild(ctx.guild).model.set(model)
         await ctx.send(f"Model set to **{model}**.")
 
+    @assistant.command(name="models")
+    async def assistant_models(self, ctx: commands.Context):
+        """List available models sorted by cost (cheapest first)."""
+        # Pricing per 1M tokens (EUR) from GreenPT docs
+        MODEL_PRICING = {
+            "deepseek-v4-flash-0731": {"input": 0.14, "output": 0.35},
+            "minimax-m2.5": {"input": 0.17, "output": 0.99},
+            "mistral-small-3.2-24b-instruct-2506": {"input": 0.20, "output": 0.40},
+            "gpt-oss-120b": {"input": 0.20, "output": 0.70},
+            "voxtral-small-24b-2507": {"input": 0.20, "output": 0.45},
+            "green-l": {"input": 0.25, "output": 0.80},
+            "green-l-raw": {"input": 0.25, "output": 0.80},
+            "qwen3-coder-30b-a3b-instruct": {"input": 0.25, "output": 0.95},
+            "gemma-3-27b-it": {"input": 0.30, "output": 0.60},
+            "qwen3.6-35b-a3b": {"input": 0.30, "output": 1.80},
+            "green-r": {"input": 0.35, "output": 0.95},
+            "green-r-raw": {"input": 0.35, "output": 0.95},
+            "gemma4": {"input": 0.50, "output": 1.50},
+            "devstral-2-123b-instruct-2512": {"input": 0.50, "output": 2.40},
+            "kimi-k2.6": {"input": 0.66, "output": 3.75},
+            "qwen3.5-397b-a17b": {"input": 0.70, "output": 4.35},
+            "kimi-k2.7-code": {"input": 0.79, "output": 3.85},
+            "qwen3-235b-a22b-instruct-2507": {"input": 0.90, "output": 2.70},
+            "glm-5.2": {"input": 1.10, "output": 4.40},
+            "llama-3.3-70b-instruct": {"input": 1.10, "output": 1.10},
+            "mistral-medium-3.5-128b": {"input": 1.80, "output": 9.00},
+            "kimi-k3": {"input": 3.30, "output": 16.50},
+        }
+
+        api_key = await self.config.api_key()
+        if not api_key or not self.session:
+            await ctx.send("API key not set. Use `[p]assistant setkey` first.")
+            return
+
+        # Fetch live model list from API
+        try:
+            async with self.session.get(
+                f"{self.BASE_URL}/models",
+                headers={"Authorization": f"Bearer {api_key}"},
+                timeout=aiohttp.ClientTimeout(total=15),
+            ) as resp:
+                if resp.status != 200:
+                    await ctx.send(f"Failed to fetch models (HTTP {resp.status}).")
+                    return
+                data = await resp.json()
+        except Exception as exc:
+            await ctx.send(f"Request failed: {exc}")
+            return
+
+        models = data.get("data", [])
+        if not models:
+            await ctx.send("No models returned by the API.")
+            return
+
+        # Build list with pricing, sorted by combined cost (input + output)
+        current_model = await self.config.guild(ctx.guild).model()
+        lines = []
+        model_entries = []
+        for m in models:
+            model_id = m["id"]
+            pricing = MODEL_PRICING.get(model_id)
+            if pricing:
+                total = pricing["input"] + pricing["output"]
+                model_entries.append((total, model_id, pricing))
+            else:
+                # Unknown pricing, put at the end
+                model_entries.append((9999, model_id, None))
+
+        model_entries.sort(key=lambda x: x[0])
+
+        for total, model_id, pricing in model_entries:
+            marker = " ◀" if model_id == current_model else ""
+            if pricing:
+                lines.append(
+                    f"`{model_id}` — €{pricing['input']:.2f} in / €{pricing['output']:.2f} out{marker}"
+                )
+            else:
+                lines.append(f"`{model_id}` — pricing unknown{marker}")
+
+        embed = discord.Embed(
+            title="Available Models (cheapest first)",
+            description="\n".join(lines),
+            color=await ctx.embed_color(),
+        )
+        embed.set_footer(text="Prices in EUR per 1M tokens. ◀ = current model.")
+        await ctx.send(embed=embed)
+
     @assistant.command(name="temp")
     async def assistant_temp(self, ctx: commands.Context, temperature: float):
         """Set the temperature (0.0 - 2.0)."""