LiveCodeBench benchmark

265 AI models ranked on LiveCodeBench. Each row shows the model’s score next to its cheapest API price per 1M tokens where it has one, so you can weigh quality against cost.

Top model today

Gemini 3 Pro (Google) leads at 92%, from $2.00 per 1M input.

Data via arXiv:2403.07974

Models ranked

265

on this benchmark

Top score

92%

current leader

Type

Single eval

benchmark

Updated

15 July 2026

last source fetch

Cost vs LiveCodeBench

160 priced models
Most attractive quadrant (cheap + high)
= reasoning model (extended thinking)

Each labelled name links to that model’s page. Curated to the cost-efficiency frontier, top scorers and cheapest so labels stay readable; the full field is in the leaderboard table below (and via “select all”). Reasoning models (extended “thinking” before answering) carry a ring around the dot.

Models ranked on LiveCodeBench

highest score first · 265 models
Maker
#
1Gemini 3 ProGoogle92%$2.00$12.00
2Gemini 3 FlashGoogle91%reasoning: truen/an/a
3DeepSeek V3 2 SpecialeDeepSeek90%$0.58$1.68
4GLM 4.7Zhipu89%$0.40$1.50
5GPT-5.2OpenAI89%medium$1.75$14.00
6gpt-oss-120bOpenAI88%$0.05$0.25
7Claude Opus 4.5Anthropic87%reasoning: true$5.00$25.00
8GPT-5.1OpenAI87%$1.25$10.00
9Mimo V2 FlashXiaomi87%reasoning: true$0.10$0.30
10DeepSeek V3.2DeepSeek86%reasoning: true$0.27$0.40
11o4 MiniOpenAI86%$1.00$4.00
12Kimi K2 ThinkingMoonshot85%$0.60$1.20
13GPT-5.1 CodexOpenAI85%n/an/a
14GPT-5OpenAI85%$1.25$10.00
15GPT-5 CodexOpenAI84%$1.25$10.00
16GPT-5 MiniOpenAI84%$0.25$2.00
17GPT-5.1 Codex miniOpenAI84%n/an/a
18Grok 4 Fast ReasoningxAI83%$0.20$0.50
19MiniMax M2MiniMax83%$0.26$1.02
20Grok 4.1 Fast ReasoningxAI82%$0.20$0.50
21Grok 4xAI82%$3.00$15.00
22ERNIE 5.0 Thinking PreviewBaidu81%n/an/a
23MiniMax M2.1MiniMax81%$0.27$1.20
24o3OpenAI81%$2.00$8.00
25Apriel-v1.6-15B-ThinkerServicenow81%n/an/a
26Gemini 2.5 ProGoogle80%$1.25$10.00
27DeepSeek V3.1 TerminusDeepSeek80%reasoning: true$0.27$1.00
28GPT-5 NanoOpenAI79%$0.05$0.40
29DeepSeek V3.2 ExpDeepSeek79%n/an/a
30Qwen3 235B A22b Instruct 2507Alibaba79%reasoning: true$0.09$0.58
31DeepSeek V3 1DeepSeek78%reasoning: true$0.27$1.00
32Qwen3 Next 80B A3bAlibaba78%reasoning: true$0.15$1.20
33Gemini 2.5 Pro (Mar)Google78%n/an/a
34gpt-oss-20bOpenAI78%$0.01$0.07
35INTELLECT-3Prime-intellect78%n/an/a
36R1DeepSeek77%$0.40$0.40
37Gemini 2.5 Pro (May)Google77%n/an/a
38K-EXAONELg-ai-research77%n/an/a
39Qwen3 MaxAlibaba77%$2.11$8.45
40Doubao Seed CodeBytedance-seed77%n/an/a
41Seed-OSS-36B-InstructBytedance-seed77%n/an/a
42Magistral Medium 2509Mistral75%$2.00$5.00
43EXAONE 4.0 32BLg-ai-research75%reasoning: truen/an/a
44Nvidia Nemotron Nano 3 30BNVIDIA74%reasoning: true$0.06$0.24
45GLM 4.5Zhipu74%$0.40$1.60
46Llama Nemotron Super 49B v1.5NVIDIA74%reasoning: truen/an/a
47o3 Mini HighOpenAI73%$1.10$4.40
48Nova 2.0 Pro PreviewAmazon73%mediumn/an/a
49Apriel-v1.5-15B-ThinkerServicenow73%n/an/a
50Nvidia Nemotron Nano 9BNVIDIA72%reasoning: true$0.04$0.16
150 of 265
Page 1 of 6
Prices are on-demand list rates. Enterprise commitments can be materially lower. Cost per task uses best-effort token usage.

Frequently asked questions

What is the LiveCodeBench benchmark?

A contamination-free coding benchmark that continuously collects new competitive-programming problems from LeetCode, AtCoder and Codeforces, tagging each with a release date so models are tested only on problems published after their training cutoff. It scores code generation, self-repair, execution and test-output prediction.

Which AI model scores highest on LiveCodeBench?

Gemini 3 Pro (Google) leads with 92%, from $2.00 per 1M input tokens.

How many models are ranked on LiveCodeBench?

265 models carry a LiveCodeBench score. Scores come from Artificial Analysis; where a model has a live API price it is verified against LiteLLM and OpenRouter, and models with no current price are listed without one.

Other benchmarks

compare the same models on a different eval
Benchmark scores via Artificial Analysis · pricing verified against LiteLLM + OpenRouter · refreshed daily · updated 15 July 2026

Every weekday

AI moves fast. Here's your debrief.

News, analysis, tools, and more.

For people who build with AI