Suggested LLMs
✨ Default Model – DeepSeek V3 (0324)
The central model responsible for generating spoken lines, in-character dialogue, combat banter, memory processing, and action decisions. Acts like a Dungeon Master, guiding the scene and ensuring the world feels alive. This is the workhorse model in SkyrimNet — chosen for its balanced creativity, reasoning skills, and cost efficiency.
Why it works well here:
- Handles long, context-heavy prompts without losing detail.
- Maintains consistent personality and tone for roleplay.
- Strong reasoning for action selection and combat chatter.
- Cost-effective enough to be used frequently for all main gameplay interactions.
Alternative Models:
DeepSeek V3.2, is the more recent iteration of Deepseek V3, will typically outperform v3 0234, its very capable and balanced, usually a better alternative to similar priced models.
meta-llama/llama-4-maverick - a similarly priced model to deepseek, with a more "helpful" general tone. It is more consistent, taking less creative risks, which makes him feel somewhat more predictable than deepseek. If using Maverick you may want to raise the model temperature to 1, due to this. It is image capable, and one of the best for this price range. This aspect makes him worthwhile to appear on this list, since llm vision integrations will come to skyrimnet very soon. ( vision isnt available yet, as of Beta5)
anthropic/claude-4.5-sonnet - a high quality roleplay model, offering realistic nuance, great prose and overall top notch experience. The price is its major shortcoming, with costs averaging $1 per hour of play (just this model alone), so have that in mind. It also tends to be verbose, though that can be managed through prompting.
🔄 Model Rotation (Only for Dialogue Model)
SkyrimNet supports model rotation for dialogue generation.
You can list multiple models in the model name field, separated by commas. After each conversation turn, the system automatically switches to the next model in the list (round-robin).
- Applies only to the
Default(Conversation) model type - Adds response variety
- Lets stronger models periodically keep weaker ones on-track
Example: openrouter/deepseek-v3.2, anthropic/claude-3.7-sonnet, anthropic/claude-4.5
Each NPC response advances to the next model, then loops back to the start.
🧠 Memory Generation Model – DeepSeek V3 (0324)
Handles memory creation, summarization, and storage.
Using the same model as dialogue ensures memories are phrased and “thought” in the same style as NPC speech, keeping personalities aligned, though having the same model is not crucial.
** Memory Creation**
- Summarizes recent event streams into first-person memories.
- Assigns:
importance_scoreemotiontagstype(TRAUMA, EXPERIENCE, etc.)
- Used later for recall, relationships, and mood shaping.
📍 View in-game under: UI > Memories
Alternative Models:
- Qwen/qwen3-235b-a22b-2507 instruct , generates detailed, vivid memories, but still very grounded on the dialogue and events that produced them. Very good quality for the cost.
🧬 Character Profile Generation Model – Claude 4.5 Sonnet current: deepseek/deepseek-r1-0528
Generates and updates NPC identity data.
Why Sonnet 4.5 here:
- Excellent at structured, consistent character building.
- Rich vocabulary for detailed backstories and personality profiles.
- Maintains thematic consistency while updating profiles over time.
** Profile Creation**
- Auto-generates profiles for modded or uninitialized NPCs.
- Includes:
- Goals
- Personality traits
- Background
- Relationship tendencies
** Dynamic Updates**
- Modifies profiles based on:
- Important memories
- Emotional shifts
- Allows characters to evolve naturally.
Alternative Models:
Several, less capable models are often unable to comply with the prompt Json instructions, so you should be careful what model to choose for this use.
-Deepseek V3 0324 , can be a cheaper alternative than sonnet.
-Qwen/qwen3-235b-a22b-2507 instruct produced very good results, costing only a small fraction of Sonnet
🎭 Action Evaluation Model – current: x-ai/grok-4.1-fast
Decides what an NPC does after speaking or reacting to an event.
Using the same model for actions and dialogue ensures NPC behavior matches what they say.
** Action Selection**
- Picks gameplay actions that follow logically from dialogue.
- Examples:
FollowPlayer,SlapTarget,PickUpItem.
** Gesture Generation**
- Chooses expressive gestures and animations to match tone.
- Examples: nodding, shrugging, dramatic motions.
Alternative Models:
DeepSeek V3 (0324), is a cheap, competent llm, capable of dealing with the context and action parameters.
🧪 Meta Evaluation Model – current x-ai/grok-4.1-fast
Performs high-frequency, small tasks that keep scenes running smoothly.
Why grok 4.1 here:
- Extremely fast and cost-efficient for micro-decisions.
- Good at short context analysis without hallucination.
- Perfect for frequent updates like turn-taking and mood tracking.
😐 Mood Evaluation
- Adjusts NPC emotional state from:
- Dialogue tone
- Memories
- Player actions
- Affects:
- Voice tone (for higher end TTS)
- Facial expressions
- Decision-making
👥 Speaking Turn Selector
- Chooses the next speaker based on relevance, social rules, and proximity.
- Ensures realistic pacing in conversations.
Alternative Models:
Gemini 2.5 Flash
A solid alternative, it previously was the default meta llm.
🧪 Diary Generation Model – anthropic/claude-sonnet-4.5
Alternative Models:
Like for Bios generations, cheaper but still creative models will be good enough, costing much less:
-Deepseek V3 0324 , can be a cheaper alternative than sonnet. -Qwen/qwen3-235b-a22b-2507 instruct produced very good results, costing only a small fraction of Sonnet
🧪 Vison Model – qwen/qwen3-vl-8b-instruct
Alternative Models:
🧪 Agent helper Model – anthropic/claude-haiku-4.5
Alternative Models:
OpenRouter SkyrimNet data
The charts displaying show the models by token count. The ranking doesnt mean higher is better, only that has more usage, often due to the task itself, like the grok4 fast models used as Meta.
View current analytics on OpenRouter
