Skip to content

Cursor最佳AI模型

AI-Native IDE

Cursor is an AI-native code editor built on VS Code. It uses AI models for code generation, editing, and chat. Models need strong code generation, reliable instruction-following, and fast response times for interactive editing.

Last updated: 47m ago
#1
Qwen3.8 Max
Alibaba
工具评分
87
输出$/百万
$6.00
Arena Elo
1497
#2
Gemini 3.6 Flash
Google
工具评分
86
输出$/百万
$7.50
Arena Elo
1485
#3
Muse Spark 1.1
meta
工具评分
86
输出$/百万
$4.25
Arena Elo
1487

Cursor的重要因素

Function CallingStreamingJSON Mode大上下文强编程能力

Best Models for Cursor

Top 15 by tool-optimized score

LMMarketCap.com

Cursor全部模型排名(378个模型)

Scored by: benchmark performance (90%) from MMLU, GPQA, HumanEval, SWE-bench, and 15+ standardized evaluations, with capabilities and context as tiebreakers (10%).

#模型评分输出$/百万
1Qwen3.8 Max
Arena Elo: 1497
87$6.00
2Gemini 3.6 Flash
Arena Elo: 1485
86$7.50
3Muse Spark 1.1
Arena Elo: 1487
86$4.25
4Claude Fable 5 (batch)86$25.00
5Gemini 3.5 Flash
Arena Elo: 1477
86$9.00
6Claude Opus 5 (Fast)85$50.00
7Claude Opus 585$25.00
8Grok 4.5
Arena Elo: 1468
85$6.00
9Claude Fable 5
SWE-bench: 95%
85$50.00
10Claude Opus 4.8 (Fast)85$50.00
11Claude Opus 4.8 (batch)85$12.50
12Claude Opus 4.7 (Fast)85$150.00
13MiMo-V2.5-Pro
Arena Elo: 1467
85$0.870
14Claude Opus 4.7 (batch)85$12.50
15GLM 5.1
Arena Elo: 1468
85$2.99
16Gemini 3.5 Flash Lite
Arena Elo: 1459
84$2.50
17Hy3
Arena Elo: 1453
84$0.528
18Qwen3.7 Plus
Arena Elo: 1458
84$1.28
19Qwen3.6 Max Preview
Arena Elo: 1460
84$6.16
20GPT-5.5 Pro (batch)84$90.00
21GPT-5.5 (batch)84$15.00
22Kimi K2.6
Arena Elo: 1461
84$2.44
23Gemini 3.1 Pro Preview Custom Tools84$12.00
24Gemini 3.1 Pro Preview (batch)84$6.00
25GLM 5
Arena Elo: 1457
84$2.55
26Inkling
Arena Elo: 1442
83$4.05
27MiniMax M3
Arena Elo: 1445
83$1.20
28Gemma 4 31B
Arena Elo: 1451
83$0.340
29Qwen3.6 Plus
Arena Elo: 1443
83$1.95
30GPT-5.4 Pro83$180.00
31GPT-5.4 Pro (batch)83$90.00
32GPT-5.4 (batch)83$7.50
33GPT-5.3-Codex83$14.00
34Qwen3.5 397B A17B
Arena Elo: 1442
83$2.34
35Claude Opus 4.6 (batch)83$12.50
36GPT-5.2-Codex83$14.00
37GLM 4.7
Arena Elo: 1442
83$1.75
38GPT-5.2 Pro83$168.00
39GPT-5.2 Pro (batch)83$84.00
40GPT-5.2 (batch)83$7.00
41GPT-5.6 Luna Pro82$0.600
42GPT-5.6 Luna Pro (batch)82$0.600
43GPT-5.6 Luna82$0.600
44GPT-5.6 Luna (batch)82$0.600
45GPT-5.6 Terra Pro82$6.00
46GPT-5.6 Terra Pro (batch)82$6.00
47GPT-5.6 Terra82$6.00
48GPT-5.6 Terra (batch)82$6.00
49GPT-5.6 Sol Pro82$30.00
50GPT-5.6 Sol Pro (batch)82$15.00
Frequently Asked Questions

Based on our analysis of coding benchmarks, capability matching, and pricing, Qwen3.8 Max currently ranks #1 for Cursor. Rankings are rebuilt as benchmark, pricing, and provider data refresh.

We score models using benchmark performance (90%) from LMArena, HumanEval, SWE-bench, MMLU, and 15+ standardized evaluations. Capabilities and context serve as tiebreakers (10%). Only models with the capabilities Cursor needs are included in the tool-specific rankings.

We currently track 378 AI models compatible with Cursor. This includes models from OpenAI, Anthropic, Google, DeepSeek, and other providers accessible via API.

Many open-source models are compatible with Cursor through API providers like OpenRouter, Together AI, and Groq. Check our rankings to see which open-source models perform best.

Rankings refresh whenever the underlying benchmark, pricing, and catalog sources refresh. That means some signals update faster than others, and the page reflects the latest verified source data available.

Best AI Models for Cursor - 2026 Rankings | LM Market Cap