Deterministic Stack Engine

Best AI Stack for lowest-latency chat

TTFT and tokens/sec first. For support widgets and voice-adjacent loops.

Objective Weight Distribution

Elo:35%TTFT:25%Speed:20%Out $:20%

Recommended PickScore 80.9 / 100

Gemini 3.7 Flash

Google · Closed flagship

elo p76ttft ms p99tokens per sec p99output price per m p49

Output: $1.875/1MSWE-bench: 72.4%

View full model fact sheet

Full Stack Leaderboard

Ranked alternatives optimized for lowest-latency chat based on objective benchmark weighting.

#2GPT-5.6 Luna(OpenAI)

Score 75.3

eloP49ttft msP98tokens per secP95output price per mP73

Rank #2 in stackvs #1

#3Gemini 3.6 Flash(Google)

Score 74.2

eloP68ttft msP96tokens per secP98output price per mP34

Rank #3 in stackvs #1

#4Gemini 3 Flash(Google)

Score 73.2

eloP66ttft msP94tokens per secP96output price per mP37

Rank #4 in stackvs #1

#5DeepSeek V4 Flash(DeepSeek)

Score 72.9

eloP61ttft msP75tokens per secP81output price per mP83

Rank #5 in stackvs #1

#6Yi-Lightning(01.AI)

Score 72.3

eloP55ttft msP80tokens per secP76output price per mP89

Rank #6 in stackvs #1

#7GPT-5.6 Terra(OpenAI)

Score 67.4

eloP81ttft msP77tokens per secP74output price per mP25

Rank #7 in stackvs #1

#8Doubao Lite 1.5(ByteDance)

Score 67.3

eloP22ttft msP91tokens per secP92output price per mP92

Rank #8 in stackvs #1

#9Grok 4.6(xAI)

Score 67.0

eloP94ttft msP61tokens per secP69output price per mP25

Rank #9 in stackvs #1

#10Seed 2.1 Turbo(ByteDance)

Score 66.5

eloP54ttft msP88tokens per secP89output price per mP39

Rank #10 in stackvs #1