Find the models that fit your workload by capability, task economics, speed, context, and provider availability.
8 results · sorted by intelligence
| # | Details | |||||||
|---|---|---|---|---|---|---|---|---|
| 1 | Gemma 4 31B (Non-reasoning)Google · Parasail | 21.8 | 33.2 | 11 | $0.034 | 15s | 262K | |
| 2 | Gemma 4 31B (Non-reasoning)Google · Cerebras | 21.8 | 33.2 | 11 | $0.034 | 15s | 131K | |
| 3 | Gemma 4 31B (Non-reasoning)Google · FriendliAI | 21.8 | 33.2 | 11 | $0.034 | 15s | 262K | |
| 4 | Gemma 4 31B (Non-reasoning)Google · Together.ai · FP8 | 21.8 | 33.2 | 11 | $0.034 | 15s | 262K | |
| 5 | Gemma 4 31B (Non-reasoning)Google · SiliconFlow · FP8 | 21.8 | 33.2 | 11 | $0.034 | 15s | 262K | |
| 6 | Gemma 4 31B (Non-reasoning)Google · Deepinfra · FP8 | 21.8 | 33.2 | 11 | $0.034 | 15s | 262K | |
| 7 | Gemma 4 31B (Non-reasoning)Google · SambaNova | 21.8 | 33.2 | 11 | $0.034 | 15s | 131K | |
| 8 | Gemma 4 31B (Non-reasoning)Google · Novita | 21.8 | 33.2 | 11 | $0.034 | 15s | 262K |
Frontier models often cluster tightly on raw intelligence. Cost per task, response time, provider coverage, and context length usually create the real shortlist.
Use Explore to find the shape of the market, then move into Compare or use the Agentic Fit Finder when task-level reliability and cost matter more than chat quality alone.
Claude Fable 5 leads the overall quality ranking right now. The best model for you depends on your use case — coding, cost, speed, and context length all shift the answer.
Sort by Quality Index for overall strength, then filter by price, speed, or context window to match your constraints. Move to the Compare page to put 2–4 finalists head to head.
Open-weight models like DeepSeek, Qwen, and Llama often lead on quality-to-price. For agentic workflows, cost per task is usually a better filter than token price alone.
Data is pulled from Artificial Analysis and refreshed automatically. New models appear as soon as they have benchmark scores and provider endpoints.
Model rankings
Browse the latest ranking pages for overall models, coding, local hardware, open source, Ollama, long context, and agentic workflows.
Current coding leaderboard using LiveCodeBench, Terminal-Bench, and SciCode.
Top open-weight models for self-hosting, Ollama, and low-cost API use.
Best local AI models by hardware tier for self-hosting on Macs, RTX GPUs, and workstations.
Coding-focused local models mapped to 8GB, 24GB, 64GB, and server-class hardware.
Ollama-first picks for coding, chat, reasoning, and low-friction local inference.
Best long-context models for large documents, codebases, and retrieval-heavy workflows.
Rankings for tool use, multi-step execution, and autonomous agent workflows.