Deterministic Stack Engine

Best AI Stack for cheap coding agents

SWE-bench first, then price. For CI bots and repo agents that cannot burn Opus prices.

Objective Weight Distribution

SWE-bench:40%Out $:35%Speed:15%Code Elo:10%

Recommended PickScore 76.4 / 100

DeepSeek V4 Flash

DeepSeek · Open weights

SWE-bench: 71.2% (Top 26%)Out $: $0.28/1M (Top 17%)Speed: 142 tok/s (Top 19%)Code Elo: 1,516 (Top 44%)

Output: $0.28/1MSWE-bench: 71.2%

View full model fact sheet

Full Stack Leaderboard

Ranked alternatives optimized for cheap coding agents based on objective benchmark weighting.

#2Gemini 3.7 Flash(Google)

Score 71.5

SWE-bench72.4%Out $$1.875/1MSpeed215 tok/sCode Elo1,546

Rank #2 in stackvs #1

#3MiniMax M2.5(MiniMax)

Score 71.5

SWE-bench75.8%Out $$0.9/1MSpeed90 tok/sCode Elo1,536

Rank #3 in stackvs #1

#4GLM-5.3(Zhipu)

Score 69.7

SWE-bench76.4%Out $$1.8/1MSpeed90 tok/sCode Elo1,586

Rank #4 in stackvs #1

#5Gemini 3 Flash(Google)

Score 67.6

SWE-bench75.8%Out $$3/1MSpeed190 tok/sCode Elo1,510

Rank #5 in stackvs #1

#6OpenAI: o3 Mini(OpenAI)

Score 66.0

SWE-bench78.5%Out $$4.4/1MSpeed92 tok/sCode Elo1,585

Rank #6 in stackvs #1

#7Gemini 2.0 Flash Thinking(Google)

Score 66.0

SWE-bench66.8%Out $$0.4/1MSpeed115 tok/sCode Elo1,515

Rank #7 in stackvs #1

#8Yi-Lightning(01.AI)

Score 65.8

SWE-bench63.4%Out $$0.14/1MSpeed130 tok/sCode Elo1,480

Rank #8 in stackvs #1

#9Gemini 3.6 Flash(Google)

Score 61.0

SWE-bench70.8%Out $$3.75/1MSpeed198 tok/sCode Elo1,514

Rank #9 in stackvs #1

#10Gemini 3.6 Pro(Google)

Score 60.5

SWE-bench74.6%Out $$5/1MSpeed102 tok/sCode Elo1,566

Rank #10 in stackvs #1

Deterministic Stack Engine

Best AI Stack for cheap coding agents

SWE-bench first, then price. For CI bots and repo agents that cannot burn Opus prices.

Objective Weight Distribution

SWE-bench:40%Out $:35%Speed:15%Code Elo:10%

Recommended PickScore 76.4 / 100

DeepSeek V4 Flash

DeepSeek · Open weights

SWE-bench: 71.2% (Top 26%)Out $: $0.28/1M (Top 17%)Speed: 142 tok/s (Top 19%)Code Elo: 1,516 (Top 44%)

Output: $0.28/1MSWE-bench: 71.2%

View full model fact sheet

Full Stack Leaderboard

Ranked alternatives optimized for cheap coding agents based on objective benchmark weighting.

#2Gemini 3.7 Flash(Google)

Score 71.5

SWE-bench72.4%Out $$1.875/1MSpeed215 tok/sCode Elo1,546

Rank #2 in stackvs #1

#3MiniMax M2.5(MiniMax)

Score 71.5

SWE-bench75.8%Out $$0.9/1MSpeed90 tok/sCode Elo1,536

Rank #3 in stackvs #1

#4GLM-5.3(Zhipu)

Score 69.7

SWE-bench76.4%Out $$1.8/1MSpeed90 tok/sCode Elo1,586

Rank #4 in stackvs #1

#5Gemini 3 Flash(Google)

Score 67.6

SWE-bench75.8%Out $$3/1MSpeed190 tok/sCode Elo1,510

Rank #5 in stackvs #1

#6OpenAI: o3 Mini(OpenAI)

Score 66.0

SWE-bench78.5%Out $$4.4/1MSpeed92 tok/sCode Elo1,585

Rank #6 in stackvs #1

#7Gemini 2.0 Flash Thinking(Google)

Score 66.0

SWE-bench66.8%Out $$0.4/1MSpeed115 tok/sCode Elo1,515

Rank #7 in stackvs #1

#8Yi-Lightning(01.AI)

Score 65.8

SWE-bench63.4%Out $$0.14/1MSpeed130 tok/sCode Elo1,480

Rank #8 in stackvs #1

#9Gemini 3.6 Flash(Google)

Score 61.0

SWE-bench70.8%Out $$3.75/1MSpeed198 tok/sCode Elo1,514

Rank #9 in stackvs #1

#10Gemini 3.6 Pro(Google)

Score 60.5

SWE-bench74.6%Out $$5/1MSpeed102 tok/sCode Elo1,566

Rank #10 in stackvs #1