CompareLLM.ai
Live
Leaderboard
Models
Compare
Best of & Stacks
Research & News
…
CompareLLM.ai
Precision Benchmarks

Programmatic, dated AI model benchmarks, head-to-head comparisons, and Stack Engine presets.

Daily ingest · 06:00 UTC

Analytics & Benchmarks

  • AI Model Leaderboard
  • Head-to-Head Compare Hub
  • Models Directory
  • Stack Engine Presets
  • Frontier Models
  • Open Weights Catalog

Guides & Intent Lists

  • Best LLM Lists (2026)
  • Best Coding LLM
  • Best Cheap LLM
  • Fastest Low-Latency LLM
  • Claude vs GPT Benchmark
  • What is Elo?
  • Methodology Guides
  • News & Dispatches

Transparency & API

  • Evaluation Methodology
  • Benchmark Changelog
  • Public JSON API
  • llms.txt Specification
  • Privacy Policy
  • Sign In / Account

© 2026 CompareLLM. Public benchmark data aggregated from Arena Elo, LiveBench, SWE-bench & OpenRouter.

Every score has a dated snapshot.

Theme:
Currency:
  1. Home
  2. Best lists
  3. cheapest coding llm
Search Intent · cheapest coding llm

Cheapest coding LLM in 2026

Repo agents ranked with SWE-bench first under a mid-tier budget. For CI bots that cannot burn Opus prices.

Quick answer

DeepSeek V4 Flash is the current #1 for “cheapest coding llm” on this dated Stack Engine mix. SWE-bench first, then price. For CI bots and repo agents that cannot burn Opus prices. Weights: SWE-bench 40%, Out $ 35%, Speed 15%, Code Elo 10%.
Weights:SWE-bench 40%Out $ 35%Speed 15%Code Elo 10%
Current #1 Ranked 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%)
SWE-bench: 71.2%Elo: 1,490
View full model fact sheet

Complete Ranked Category List

Models ranked by verified benchmark weights across SWE-bench and coding preference evaluations.

#1DeepSeek V4 Flash(DeepSeek)
Score 76.4
SWE-bench: 71.2% (Top26%)Out $: $0.28/1M (Top17%)Speed: 142 tok/s (Top19%)Code Elo: 1,516 (Top44%)
🏆 #1 Overall Leader in Category
#2Gemini 3.7 Flash(Google)
Score 71.5
SWE-bench: 72.4% (Top21%)Out $: $1.875/1M (Top50%)Speed: 215 tok/s (Top1%)Code Elo: 1,546 (Top24%)
Rank #2vs #1
#3MiniMax M2.5(MiniMax)
Score 71.5
SWE-bench: 75.8% (Top12%)Out $: $0.9/1M (Top35%)Speed: 90 tok/s (Top56%)Code Elo: 1,536 (Top31%)
Rank #3vs #1
#4GLM-5.3(Zhipu)
Score 69.7
SWE-bench: 76.4% (Top9%)Out $: $1.8/1M (Top50%)Speed: 90 tok/s (Top56%)Code Elo: 1,586 (Top8%)
Rank #4vs #1
#5Gemini 3 Flash(Google)
Score 67.6
SWE-bench: 75.8% (Top12%)Out $: $3/1M (Top63%)Speed: 190 tok/s (Top4%)Code Elo: 1,510 (Top50%)
Rank #5vs #1
#6OpenAI: o3 Mini(OpenAI)
Score 66.0
SWE-bench: 78.5% (Top3%)Out $: $4.4/1M (Top69%)Speed: 92 tok/s (Top51%)Code Elo: 1,585 (Top10%)
Rank #6vs #1
#7Gemini 2.0 Flash Thinking(Google)
Score 66.0
SWE-bench: 66.8% (Top41%)Out $: $0.4/1M (Top23%)Speed: 115 tok/s (Top33%)Code Elo: 1,515 (Top46%)
Rank #7vs #1
#8Yi-Lightning(01.AI)
Score 65.8
SWE-bench: 63.4% (Top51%)Out $: $0.14/1M (Top11%)Speed: 130 tok/s (Top24%)Code Elo: 1,480 (Top64%)
Rank #8vs #1
#9Gemini 3.6 Flash(Google)
Score 61.0
SWE-bench: 70.8% (Top28%)Out $: $3.75/1M (Top65%)Speed: 198 tok/s (Top2%)Code Elo: 1,514 (Top48%)
Rank #9vs #1
#10Gemini 3.6 Pro(Google)
Score 60.5
SWE-bench: 74.6% (Top15%)Out $: $5/1M (Top72%)Speed: 102 tok/s (Top43%)Code Elo: 1,566 (Top18%)
Rank #10vs #1
#11GPT-5.6 Luna(OpenAI)
Score 59.8
SWE-bench: 61.8% (Top56%)Out $: $0.6/1M (Top27%)Speed: 188 tok/s (Top5%)Code Elo: 1,458 (Top76%)
Rank #11vs #1
#12Qwen 2.5 Coder 32B(Alibaba)
Score 59.2
SWE-bench: 65.2% (Top46%)Out $: $0.72/1M (Top31%)Speed: 110 tok/s (Top39%)Code Elo: 1,495 (Top57%)
Rank #12vs #1

Frequently asked questions

Plain-English methodology and leaderboard answers

DeepSeek V4 Flash is the current #1 on this list. Rankings move when daily ingest updates SWE-bench, Elo, price, or latency.

What is preference Elo? · How rankings update