15 listings — 9 routed · 6 market comparisons

Claude Fable 5FrontierRecommended

Frontier agentic reasoning, long-context work, and difficult engineering.

by Anthropic·AA index 60 · Core fit 64 at max effort·$2.75 per benchmark task
$10 / $50 per 1M60
Kimi K3FrontierRecommended

Long-context coding, knowledge work, and agentic workflows.

by Moonshot AI·AA index 57 · Core fit 61 at max effort·$0.95 per benchmark task
$3 / $15 per 1M57
Grok 4.5StrongRecommendedCore default

Core's default for strong general reasoning and agent work.

by Grok·AA index 54 · Core fit 59 at high effort·$0.31 per benchmark task
$2 / $6 per 1M54
Claude Opus 4.8Strong

Deep engineering and difficult multi-step work.

by Anthropic·AA index 56 · Core fit 58 at max effort·$1.80 per benchmark task
$5 / $25 per 1M56
GPT-5.6 SolStrong

OpenAI's highest-quality general agent model in Core.

by OpenAI·AA index 59 · Core fit 56 at max effort·$1.04 per benchmark task
$5 / $30 per 1M59
Claude Sonnet 5Strong

Strong daily agent quality with balanced speed and cost.

by Anthropic·AA index 53 · Core fit 55 at max effort·$1.52 per benchmark task
$2 / $10 per 1M53
Muse Spark 1.1EfficientRecommended

One-million-token multimodal reasoning for everyday agent work.

by Meta·AA index 51 · Core fit 54 at xhigh effort·$0.26 per benchmark task
$1.25 / $4.25 per 1M51
GPT-5.6 TerraEfficient

Strong daily coding and knowledge work at a moderate cost.

by OpenAI·AA index 55 · Core fit 52 at max effort·$0.82 per benchmark task
$2.5 / $15 per 1M55
GLM-5.2Market comparison

A public benchmark comparison model measured at its maximum reasoning effort.

by Z.ai·AA index 51.1 at max effort·$0.32 per benchmark task
$1.4 / $4.4 per 1M51.1
Gemini 3.6 FlashMarket comparison

A public benchmark comparison model included for its measured cost-quality position.

by Google·AA index 50.1 at effort·$0.50 per benchmark task
$1.5 / $7.5 per 1M50.1
GPT-5.6 LunaEfficient

Efficient everyday agent work and quick iteration.

by OpenAI·AA index 51 · Core fit 50 at max effort·$0.21 per benchmark task
$1 / $6 per 1M51
MiniMax-M3Market comparison

A public benchmark comparison model included for its measured cost-quality position.

by MiniMax·AA index 44.4 at effort·$0.12 per benchmark task
$0.3 / $1.2 per 1M44.4
DeepSeek V4 ProMarket comparison

A public benchmark comparison model measured at its maximum reasoning effort.

by DeepSeek·AA index 44.3 at max effort·$0.04 per benchmark task
$0.435 / $0.87 per 1M44.3
Nemotron 3 UltraMarket comparison

A public benchmark comparison model included for its measured cost-quality position.

by NVIDIA·AA index 37.8 at effort·$0.24 per benchmark task
$0.675 / $2.675 per 1M37.8
gpt-oss-120bMarket comparison

A public benchmark comparison model measured at high reasoning effort.

by OpenAI·AA index 23.8 at high effort·$0.06 per benchmark task
$0.15 / $0.6 per 1M23.8