User questions cost-effectiveness of Sol vs Terra models for routing
A Reddit user questions the cost-effectiveness of using Sol models (double the cost of Terra) for tough tasks, given that Terra is considered roughly equal to Sonnet. The user suggests using Terra High for most tasks and Sol High only for the hardest tasks, challenging a recommendation to use Sol Medium as the main model.
Entities
Related
User suggests SOL add intelligence tier recommendation based on task complexity
A Reddit user reflects that most tasks do not require the highest SOL intelligence tiers (High/Extra/Ultra) and suggests the platform add a feature to recommend an appropriate intelligence level per task, with justification. The user notes that overly high intelligence can be counterproductive for simple tasks.
User seeks advice on adding second cheap GPU to run larger local models
A Reddit user running Gemma 4 26B-A4B at 12-15 t/s on an RTX 3060 (12 GB) finds the model insufficiently intelligent and wants to upgrade to a 31B model, which runs at only 1.5 t/s. They ask the community about the benefits of adding a second cheap GPU to improve performance for local LLM inference.
Community debate: MoE vs dense models — Qwen 3.5 122B example
A Reddit post challenges the common sentiment that a 122B MoE model with 10B active parameters is equivalent to a dense 10B model, arguing that router effectiveness makes MoE more capable. The post questions why providers would release MoE models if they offered no advantage over dense models.
Community urges patience with new models after 48 hours
A Reddit user reminds the community that new models have only been out for 48 hours, and that different models suit different tasks and skill levels. They caution against accepting premature expert opinions on which model to use.
Users report Claude Code + Fable outperforms Sol + Terra in multi-agent coding workflows
Reddit users report that combining Claude Code with Fable for multi-agent coding tasks outperforms Sol + Terra, while being relatively cheap. The approach uses Fable for orchestration and Opus 4.8 for subagent work, burning fewer tokens than a full Fable ultracode setup. Users also note that small/medium coding tasks are nearly solved, citing recent AtCoder competition results.
