Models — Fronset Benchmark
Per-model performance across all capabilities.
One page per LLM in the benchmark — its performance across every task, where it’s the cheapest qualifier at the default quality bar, and which categories it does and doesn’t qualify on.
- Claude Haiku 5.5
Good enough on 17/54 tasks at the 90% bar. Best value on 4 tasks. Best fit for: Infrastructure & Utility, Long-form Content Generation, Social & Promotional Content. Doesn't qualify on any: Financial
- Claude Opus 5.5
Good enough on 33/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Relevance, Classificatio
- Claude Sonnet 5.5
Good enough on 24/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Long-form Content Generation, Social & Promotional Content.
- DeepSeek V4.1 Flash
Good enough on 18/54 tasks at the 90% bar. Best value on 5 tasks. Best fit for: Financial Analysis & Trading Decisions, Content Summarization & Synthesis.
- Gemini 3.5 Flash Lite
Good enough on 7/54 tasks at the 90% bar. Best value on 0 tasks. Doesn't qualify on any: Structured Data & Fact Extraction, Financial Analysis & Trading Decisions, Long-form Content Generation, Topic
- Gemini 3.8 Flash
Good enough on 24/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Infrastructure & Utility, Content Summarization & Synthesis.
- GLM-5.3
Good enough on 27/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Relevance, Classificatio
- GLM-5.3 Flash
Good enough on 27/54 tasks at the 90% bar. Best value on 3 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Relevance, Classificatio
- GPT-6 Luna
Good enough on 24/54 tasks at the 90% bar. Best value on 15 tasks. Best fit for: Infrastructure & Utility, Content Summarization & Synthesis.
- GPT-6 Sol
Good enough on 28/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Social & Promotional Content, Content Summarization &
- Grok 4.7
Good enough on 29/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Relevance, Classification & Matching, Social & Promoti
- Meta Muse Spark 1.3
Good enough on 34/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Relevance, Classification & Matching, Content Summariz
- MiniMax M3
Good enough on 23/54 tasks at the 90% bar. Best value on 5 tasks. Best fit for: Long-form Content Generation.
- Moonshot Kimi K3
Good enough on 37/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Structured Data & Fact Extraction, Financial Analysis & Trading Decisions, Infrastructure & Utility, Relevance, Classifi
- NVIDIA Nemotron 3.5 Lightning
Good enough on 2/54 tasks at the 90% bar. Best value on 0 tasks. Doesn't qualify on any: Financial Analysis & Trading Decisions, Infrastructure & Utility, Content Summarization & Synthesis.
- NVIDIA Nemotron-3 Nano 30B-A3B
Good enough on 2/54 tasks at the 90% bar. Best value on 0 tasks. Doesn't qualify on any: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Social & Promot
- NVIDIA Nemotron-3 Super 120B
Good enough on 6/54 tasks at the 90% bar. Best value on 0 tasks. Doesn't qualify on any: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Topic Organizat
- NVIDIA Nemotron-3 Ultra 550B
Good enough on 12/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Infrastructure & Utility, Topic Organization & Clustering. Doesn't qualify on any: Financial Analysis & Trading Decision
- Qwen 3.7 Plus
Good enough on 23/54 tasks at the 90% bar. Best value on 3 tasks. Best fit for: Social & Promotional Content.
- Qwen 3.8 Flash
Good enough on 5/54 tasks at the 90% bar. Best value on 0 tasks. Doesn't qualify on any: Financial Analysis & Trading Decisions, Infrastructure & Utility, Content Summarization & Synthesis, Topic Orga
- Qwen 3.8 Max
Good enough on 20/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Content Summarization & Synthesis.
- Tencent Hy3
Good enough on 16/54 tasks at the 90% bar. Best value on 1 task. Best fit for: Topic Organization & Clustering. Doesn't qualify on any: Financial Analysis & Trading Decisions.
- Tencent Hy4 Preview
Good enough on 27/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Relevance, Classificatio
- Thinking Machines Inkling
Good enough on 29/54 tasks at the 90% bar. Best value on 1 task. Best fit for: Financial Analysis & Trading Decisions, Infrastructure & Utility, Long-form Content Generation, Social & Promotional Cont
- Thinking Machines Inkling Small
Good enough on 27/54 tasks at the 90% bar. Best value on 3 tasks. Best fit for: Long-form Content Generation, Relevance, Classification & Matching, Social & Promotional Content.
- Xiaomi MiMo V2.6 Flash
Good enough on 16/54 tasks at the 90% bar. Best value on 14 tasks. Best fit for: Relevance, Classification & Matching, Topic Organization & Clustering. Doesn't qualify on any: Financial Analysis & Tra
- Xiaomi MiMo V2.6 Pro
Good enough on 20/54 tasks at the 90% bar. Best value on 0 tasks. Best fit for: Infrastructure & Utility, Relevance, Classification & Matching, Social & Promotional Content, Content Summarization & Sy
Archived models
No longer on the Fronset Benchmark. Each page shows the model as of the last snapshot that carried it.
- Alibaba Qwen3.7-Flash last benchmarked
- Claude Haiku 4.5 last benchmarked
- Claude Opus 4.7 last benchmarked
- Claude Opus 4.8 last benchmarked
- Claude Opus 5 last benchmarked
- Claude Sonnet 4.6 last benchmarked
- Claude Sonnet 5 last benchmarked
- DeepSeek V4 Flash last benchmarked
- DeepSeek V4 Pro last benchmarked
- Gemini 3 Flash Preview last benchmarked
- Gemini 3.1 Flash Lite last benchmarked
- Gemini 3.1 Pro Preview last benchmarked
- Gemini 3.5 Flash last benchmarked
- Gemini 3.6 Flash last benchmarked
- Gemini 3.7 Flash last benchmarked
- GPT-5.4 Mini last benchmarked
- GPT-5.4 Nano last benchmarked
- GPT-5.5 last benchmarked
- GPT-5.6 Luna last benchmarked
- GPT-5.6 Sol last benchmarked
- GPT-5.6 Terra last benchmarked
- Grok 4.5 last benchmarked
- Grok 4.6 last benchmarked
- Haiku 4.5 last benchmarked
- Kimi K2.6 last benchmarked
- Meta Muse Spark 1.1 last benchmarked
- Meta Muse Spark 1.2 last benchmarked
- MiniMax M2.5 last benchmarked
- Qwen 3.5 Flash last benchmarked
- Qwen 3.6 Flash last benchmarked
- Qwen 3.6 Plus last benchmarked