// model_matchups

Compare LLMs side by side

Every pairing gets the same treatment: LiveBench scores across all seven categories, input and output pricing, context windows, cached-input rates, and the measured dollars behind a point of capability. Prices refresh every 15 minutes, so a comparison does not go stale the week after a price cut.

Closely matched pairs

Each of the top-scoring models against the one ranked immediately below it — the comparisons where the answer is not obvious.

anthropic vs openai

Claude Fable 5 vs GPT-5.6 Sol

Claude Fable 5 scores higher, GPT-5.6 Sol costs less — it depends on your workload.

$20.00 vs $4.00 / 1MCompare

openai vs openai

GPT-5.6 Sol vs GPT-5.5

Effectively the same quality — GPT-5.6 Sol is the cheaper way to get it.

$4.00 vs $11.25 / 1MCompare

openai vs anthropic

GPT-5.5 vs Claude Opus 5

Effectively the same quality — Claude Opus 5 is the cheaper way to get it.

$11.25 vs $10.00 / 1MCompare

anthropic vs moonshotai

Claude Opus 5 vs Kimi K3

Effectively the same quality — Kimi K3 is the cheaper way to get it.

$10.00 vs $6.00 / 1MCompare

moonshotai vs google

Kimi K3 vs Gemini 3.7 Flash

Effectively the same quality — Gemini 3.7 Flash is the cheaper way to get it.

$6.00 vs $0.750 / 1MCompare

google vs x-ai

Gemini 3.7 Flash vs SpaceXAI: Grok 4.6

Effectively the same quality — Gemini 3.7 Flash is the cheaper way to get it.

$0.750 vs $3.00 / 1MCompare

x-ai vs openai

SpaceXAI: Grok 4.6 vs GPT-5.4

Effectively the same quality — SpaceXAI: Grok 4.6 is the cheaper way to get it.

$3.00 vs $5.63 / 1MCompare

openai vs meta

GPT-5.4 vs Muse Spark 1.2

Effectively the same quality — Muse Spark 1.2 is the cheaper way to get it.

$5.63 vs $2.00 / 1MCompare

meta vs openai

Muse Spark 1.2 vs GPT-5.6 Terra

Effectively the same quality — Muse Spark 1.2 is the cheaper way to get it.

$2.00 vs $4.50 / 1MCompare

openai vs deepseek

GPT-5.6 Terra vs DeepSeek V4 Pro 0813

Effectively the same quality — DeepSeek V4 Pro 0813 is the cheaper way to get it.

$4.50 vs $1.78 / 1MCompare

deepseek vs google

DeepSeek V4 Pro 0813 vs Gemini 3.1 Pro Preview

Effectively the same quality — DeepSeek V4 Pro 0813 is the cheaper way to get it.

$1.78 vs $4.50 / 1MCompare

google vs anthropic

Gemini 3.1 Pro Preview vs Claude Opus 4.7

Effectively the same quality — Gemini 3.1 Pro Preview is the cheaper way to get it.

$4.50 vs $10.00 / 1MCompare

Top models

Ranked by overall LiveBench score, with list price and context window.

ModelOverallBlended / 1MContext
Claude Fable 5anthropic83.0$20.001M
GPT-5.6 Solopenai81.1$4.001.1M
GPT-5.5openai80.2$11.251.1M
Claude Opus 5anthropic80.1$10.001M
Kimi K3moonshotai79.2$6.001.0M
Gemini 3.7 Flashgoogle78.8$0.7501.0M
SpaceXAI: Grok 4.6x-ai78.0$3.00500K
GPT-5.4openai78.0$5.631.1M
Muse Spark 1.2meta78.0$2.001.0M
GPT-5.6 Terraopenai77.9$4.501.1M
DeepSeek V4 Pro 0813deepseek77.4$1.781.0M
Gemini 3.1 Pro Previewgoogle77.0$4.501.0M

All pairings

Every combination of the models above. Any two models in the full catalogue can be compared by URL — 415 models, 59 providers.

Claude Fable 5 vs GPT-5.6 SolClaude Fable 5 vs GPT-5.5Claude Fable 5 vs Claude Opus 5Claude Fable 5 vs Kimi K3Claude Fable 5 vs Gemini 3.7 FlashClaude Fable 5 vs SpaceXAI: Grok 4.6Claude Fable 5 vs GPT-5.4Claude Fable 5 vs Muse Spark 1.2Claude Fable 5 vs GPT-5.6 TerraClaude Fable 5 vs DeepSeek V4 Pro 0813Claude Fable 5 vs Gemini 3.1 Pro PreviewGPT-5.6 Sol vs GPT-5.5GPT-5.6 Sol vs Claude Opus 5GPT-5.6 Sol vs Kimi K3GPT-5.6 Sol vs Gemini 3.7 FlashGPT-5.6 Sol vs SpaceXAI: Grok 4.6GPT-5.6 Sol vs GPT-5.4GPT-5.6 Sol vs Muse Spark 1.2GPT-5.6 Sol vs GPT-5.6 TerraGPT-5.6 Sol vs DeepSeek V4 Pro 0813GPT-5.6 Sol vs Gemini 3.1 Pro PreviewGPT-5.5 vs Claude Opus 5GPT-5.5 vs Kimi K3GPT-5.5 vs Gemini 3.7 FlashGPT-5.5 vs SpaceXAI: Grok 4.6GPT-5.5 vs GPT-5.4GPT-5.5 vs Muse Spark 1.2GPT-5.5 vs GPT-5.6 TerraGPT-5.5 vs DeepSeek V4 Pro 0813GPT-5.5 vs Gemini 3.1 Pro PreviewClaude Opus 5 vs Kimi K3Claude Opus 5 vs Gemini 3.7 FlashClaude Opus 5 vs SpaceXAI: Grok 4.6Claude Opus 5 vs GPT-5.4Claude Opus 5 vs Muse Spark 1.2Claude Opus 5 vs GPT-5.6 TerraClaude Opus 5 vs DeepSeek V4 Pro 0813Claude Opus 5 vs Gemini 3.1 Pro PreviewKimi K3 vs Gemini 3.7 FlashKimi K3 vs SpaceXAI: Grok 4.6Kimi K3 vs GPT-5.4Kimi K3 vs Muse Spark 1.2Kimi K3 vs GPT-5.6 TerraKimi K3 vs DeepSeek V4 Pro 0813Kimi K3 vs Gemini 3.1 Pro PreviewGemini 3.7 Flash vs SpaceXAI: Grok 4.6Gemini 3.7 Flash vs GPT-5.4Gemini 3.7 Flash vs Muse Spark 1.2Gemini 3.7 Flash vs GPT-5.6 TerraGemini 3.7 Flash vs DeepSeek V4 Pro 0813Gemini 3.7 Flash vs Gemini 3.1 Pro PreviewSpaceXAI: Grok 4.6 vs GPT-5.4SpaceXAI: Grok 4.6 vs Muse Spark 1.2SpaceXAI: Grok 4.6 vs GPT-5.6 TerraSpaceXAI: Grok 4.6 vs DeepSeek V4 Pro 0813SpaceXAI: Grok 4.6 vs Gemini 3.1 Pro PreviewGPT-5.4 vs Muse Spark 1.2GPT-5.4 vs GPT-5.6 TerraGPT-5.4 vs DeepSeek V4 Pro 0813GPT-5.4 vs Gemini 3.1 Pro PreviewMuse Spark 1.2 vs GPT-5.6 TerraMuse Spark 1.2 vs DeepSeek V4 Pro 0813Muse Spark 1.2 vs Gemini 3.1 Pro PreviewGPT-5.6 Terra vs DeepSeek V4 Pro 0813GPT-5.6 Terra vs Gemini 3.1 Pro PreviewDeepSeek V4 Pro 0813 vs Gemini 3.1 Pro Preview