Skip to main content

Decision Tools

Alternative Finder

Find replacements for any LLM by price, context length, benchmark score, and deployment type.

Runs locally

Reference model

Input price$5/M
Output price$15/M
Context128K tokens
MMLU88.7
Open weightNo

Constraints

Alternatives (18)

Sorted by value score

ModelInput $/MPrice ΔContextContext ΔMMLUMMLU ΔScore
best matchMetaLlama 4 Maverickopen
$0.6-88.0%256K+100.0%85-3.7pts89
GoogleGemini 3.1 Flash
$0.5-90.0%1000K+681.3%82-6.7pts89
MetaLlama 4 Scoutopen
$0.5-90.0%512K+300.0%79-9.7pts88
DeepSeekDeepSeek V3open
$0.27-94.6%128K0.0%88.5-0.2pts86
AnthropicClaude 4 Haiku
$0.8-84.0%200K+56.3%78-10.7pts85
DeepSeekDeepSeek R1open
$0.55-89.0%128K0.0%90.8+2.1pts85
OpenAIGPT-4o Mini
$0.15-97.0%128K0.0%82-6.7pts84
MistralMistral Small 3.1open
$0.1-98.0%128K0.0%72-16.7pts81
MistralCodestral 25.01
$0.9-82.0%256K+100.0%70-18.7pts81
GoogleGemini 3.1 Pro
$2-60.0%2000K+1462.5%91.5+2.8pts81
xAIGrok 4
$2-60.0%2000K+1462.5%90+1.3pts80
OpenAIGPT-5.2 Mini
$1.2-76.0%128K0.0%84-4.7pts77
OpenAIGPT-5.4
$2.5-50.0%272K+112.5%90.5+1.8pts76
AnthropicClaude Sonnet 4.6
$3-40.0%1000K+681.3%88.5-0.2pts71
MistralMistral Large 2open
$3-40.0%128K0.0%84-4.7pts62