AICoder
2026 Live Verified

AI Coding Models Leaderboard

Aggregated mirrors of 8 official benchmark boards and real-world coding task scenarios. 100% verified, objective, and zero paid placement.

8Boards
112Models ranked
4Evaluation Tasks
November 2026Period
2026-11-15Last update
●We fetch raw rankings from 6 public benchmark providers across 8 official boards — no re-weighting, no composite score. Click "Official ↗" at the bottom of each card to go to the source.
🧩Tool Matrix⚔️VS Compare💳Plans
ADSponsored

📌 8 official boards · mirrored 1:1

Each card previews a provider's Top 5. Click "View full" for the complete board, or jump to the official page. Data is 100% from each provider — we do no aggregation or math.

LMArena Code ★
Human preference blind battle coding Elo leaderboard
Coding
Metric:Coding Elo· 70 models
  1. #1Claude Fable 5.21837.8
  2. #2Claude Opus 5.51814.7
  3. #3Claude Sonnet 5.51772.3
  4. #4Gemini 4 Carbon1768.2
  5. #5GPT-6 Astra1763.8
Official ↗
LMArena Agent
Human preference on real agentic coding sessions
Agent
Metric:Agent Score· 45 models
  1. #1Claude Fable 5.230.48
  2. #2Claude Opus 5.519.51
  3. #3Gemini 4 Carbon16.9
  4. #4Claude Sonnet 5.516.07
  5. #5Gemini 4 Hydrogen14.8
Official ↗
CursorBench
Cursor's first-party agentic coding benchmark (v4.0)
Agent
Metric:Pass rate%· 21 models
  1. #1Gemini 4 Carbon60.4%
  2. #2Claude Fable 5.260.0%
  3. #3Claude Opus 5.557.8%
  4. #4Gemini 4 Hydrogen57.8%
  5. #5Claude Sonnet 5.555.5%
Official ↗
GDPx (SemiAnalysis)
SemiAnalysis general agentic benchmark across 11 high-value domains (206 tasks)
Agent
Metric:Pass@1%· 20 models
  1. #1Claude Fable 5.216.4%
  2. #2Claude Opus 5.512.0%
  3. #3Gemini 4 Carbon10.9%
  4. #4GPT-6 Astra9.6%
  5. #5Muse Wonder Full9.0%
Official ↗
Artificial Analysis
Independent industry standard Intelligence Index
Overall
Metric:Intelligence Index· 34 models
  1. #1Claude Fable 5.264.6
  2. #2Gemini 4 Carbon60
  3. #3Claude Opus 5.559.1
  4. #4GPT-6 Astra56.9
  5. #5GPT-6.1 Sol54.6
Official ↗
Vals AI
Independent evaluations on real-world tasks in software, finance & cybersecurity
Overall
Metric:Accuracy%· 28 models
  1. #1Claude Fable 5.271.3%
  2. #2Claude Sonnet 5.570.8%
  3. #3Gemini 4 Carbon70.5%
  4. #4Claude Opus 5.570.1%
  5. #5Gemini 4 Argon69.4%
Official ↗
LiveBench
Contamination-resistant coding suite (completion, generation, JS, Python, TS)
Coding
Metric:Score%· 26 models
  1. #1GPT-6 Astra85.8%
  2. #2Claude Fable 5.285.6%
  3. #3Claude Fable 5.185.1%
  4. #4Gemini 4 Carbon84.8%
  5. #5Claude Opus 5.584.7%
Official ↗
LMArena Text
Chatbot Arena text and general capability Elo rating
Chat
Metric:Arena Elo· 99 models
  1. #1Gemini 4 Carbon1526.2
  2. #2Gemini 4 Argon1521.7
  3. #3Muse Wonder1510.9
  4. #4Claude Opus 5.51506.8
  5. #5Claude Fable 51504
Official ↗

LMArena CodeCoding

Metric: Coding Elo·70 models·fetched 2026-11-15

Open official board ↗
✨

Not sure which model fits your budget & stack?

Try 4-Way Plan Showdowns & $20/mo vs API Cost Break-Even Meter

Compare Plans & Costs↗
#ModelCoding EloValue ScoreCross-board ranksActions
#1
Claude Fable 5.2Anthropic
1837.8
9.9
L #1
C #2
G #1
A #1
V #1
L #2
L—
⚔️ VS💰Plans🚀Try
#2
Claude Opus 5.5Anthropic
1814.7
18.3
L #2
C #3
G #2
A #3
V #4
L #5
L #4
⚔️ VS💰Plans🚀Try
#3
Claude Sonnet 5.5Anthropic
1772.3
33.5
L #4
C #5
G #9
A #6
V #2
L #10
L #31
⚔️ VS💰Plans🚀Try
#4
Gemini 4 CarbonGoogle
1768.2
36
L #3
C #1
G #3
A #2
V #3
L #4
L #1
⚔️ VS💰Plans🚀Try
#5
GPT-6 AstraOpenAI
1763.8
7.1
L #7
C—
G #4
A #4
V #8
L #1
L #29
⚔️ VS💰Plans🚀Try
#6
GPT-6.1 SolOpenAI
1743.3
34
L #8
C—
G #8
A #5
V #10
L #6
L #22
⚔️ VS💰Plans🚀Try
#7
Claude Fable 5.1Anthropic
1741.9
6.6
L #6
C #7
G #6
A—V—
L #3
L #7
⚔️ VS💰Plans🚀Try
#8
Muse Wonder FullMeta
1723.5
15.5
L #11
C—
G #5
A #10
V #7
L—L—
⚔️ VS💰Plans🚀Try
#9
Gemini 4 HydrogenGoogle
1704.2
45.9
L #5
C #4
G—
A #7
V #6
L #9
L #10
⚔️ VS💰Plans🚀Try
#10
Claude Opus 5Anthropic
1690.9
4.2
L #13
C #10
G—A—V—L—
L #11
⚔️ VS💰Plans🚀Try
#11
Kimi K3.1 ProMoonshot AI
1687.2
32.4
L #16
C—G—
A #11
V #11
L #13
L #17
⚔️ VS💰Plans🚀Try
#12
Gemini 4 ArgonGoogle
1681
31.9
L #10
C #6
G #10
A #9
V #5
L #8
L #2
⚔️ VS💰Plans🚀Try
#13
GPT-6.1 TerraOpenAI
1678.1
62.6
L #9
C—
G #7
A #8
V #13
L #7
L—
⚔️ VS💰Plans🚀Try
#14
Qwen3.8-MaxQwen
1676.3
35.3
L #31
C—
G #18
A #17
V #17
L #22
L #18
⚔️ VS💰Plans🚀Try
#15
GPT-6 SolOpenAI
1672.6
33.3
L #14
C—G—A—V—L—
L #40
⚔️ VS💰Plans🚀Try
#16
ERNIE 5.5 Flash PreviewBaidu
1668
284.7
L—C—G—A—V—L—L—
⚔️ VS💰Plans🚀Try
#17
DeepSeek-V4.1-FlashDeepSeek
1663.7
227.2
L #22
C—
G #12
A #18
V #12
L #12
L #26
⚔️ VS💰Plans🚀Try
#18
Kimi K3Moonshot AI
1657.8
17.9
L #28
C—
G #17
A—
V #25
L #18
L #13
⚔️ VS💰Plans🚀Try
#19
Muse Spark 1.3Meta
1655.1
48
L #21
C #14
G #20
A #27
V #28
L #17
L #9
⚔️ VS💰Plans🚀Try
#20
Muse WonderMeta
1647.4
14.5
L #25
C #8
G #11
A #12
V #9
L #11
L #3
⚔️ VS💰Plans🚀Try
#21
Inkling 2 SmallThinking Machines
1647.1
232.7
L #18
C—G—
A #21
V #21
L—
L #66
⚔️ VS💰Plans🚀Try
#22
Grok 4.7xAI
1640.7
34.6
L #19
C #11
G #16
A #15
V #19
L #16
L #49
⚔️ VS💰Plans🚀Try
#23
Hunyuan Hy4Tencent
1631.7
86.2
L #15
C—G—
A #16
V #16
L—L—
⚔️ VS💰Plans🚀Try
#24
Claude Fable 5Anthropic
1625.4
6.7
L #12
C—G—A—V—L—
L #5
⚔️ VS💰Plans🚀Try
#25
Qwen 4 PreviewQwen
1623.7
40.2
L #38
C—G—
A #23
V—L—L—
⚔️ VS💰Plans🚀Try
#26
GLM-5.3Zhipu AI
1622.5
45.1
L #27
C #12
G #19
A #19
V #15
L—
L #20
⚔️ VS💰Plans🚀Try
#27
Grok 4.6xAI
1618.1
37.3
L #32
C #15
G—A—V—
L #20
L #41
⚔️ VS💰Plans🚀Try
#28
GPT-5.6 SolOpenAI
1615.2
10.3
L #17
C #13
G—A—V—L—
L #16
⚔️ VS💰Plans🚀Try
#29
GLM-5.3-FlashZhipu AI
1608.8
426.7
L #36
C #18
G—
A #25
V #27
L #21
L #24
⚔️ VS💰Plans🚀Try
#30
Muse EmberMeta
1608.8
524.5
L #23
C—G—
A #14
V #14
L—L—
⚔️ VS💰Plans🚀Try
#31
GLM-5.2Zhipu AI
1607.7
57
L #30
C—G—A—V—L—
L #27
⚔️ VS💰Plans🚀Try
#32
Claude Haiku 5.5Anthropic
1593.4
51.5
L #29
C #9
G #14
A #13
V #18
L #15
L #61
⚔️ VS💰Plans🚀Try
#33
Qwen3.8-27BQwen
1589.7
219
L #39
C—G—
A #30
V—
L #26
L #56
⚔️ VS💰Plans🚀Try
#34
Gemini 3.7 FlashGoogle
1588
79.9
L #34
C—G—A—V—L—
L #12
⚔️ VS💰Plans🚀Try
#35
MiniMax-M3.1-FlashMiniMax
1583.9
338.3
L #35
C—
G #15
A #22
V #23
L #19
L—
⚔️ VS💰Plans🚀Try
#36
Gemini 3.8 FlashGoogle
1583.5
70.9
L #20
C #17
G—
A #20
V #22
L #25
L #8
⚔️ VS💰Plans🚀Try
#37
DeepSeek-V4-FlashDeepSeek
1581.7
412.9
L—C—G—A—V—L—
L #54
⚔️ VS💰Plans🚀Try
#38
DeepSeek-V4-ProDeepSeek
1581
141.5
L—C—G—A—V—L—
L #36
⚔️ VS💰Plans🚀Try
#39
GPT-6 LunaOpenAI
1577.2
497.7
L #26
C—
G #13
A #24
V #26
L #23
L #53
⚔️ VS💰Plans🚀Try
#40
Mistral Large 4Mistral AI
1564.5
49.1
L #40
C—G—
A #26
V #20
L #24
L #62
⚔️ VS💰Plans🚀Try
#41
Claude Opus 4.8Anthropic
1557
14.3
L—C—G—A—V—L—
L #6
⚔️ VS💰Plans🚀Try
#42
Grok 4.5xAI
1552.6
45.7
L—C—G—A—V—L—
L #34
⚔️ VS💰Plans🚀Try
#43
Claude Sonnet 5Anthropic
1539
25.9
L #24
C #20
G—A—V—L—
L #38
⚔️ VS💰Plans🚀Try
#44
Gemini 3.6 FlashGoogle
1537.7
92
L—C—G—A—V—L—
L #19
⚔️ VS💰Plans🚀Try
#45
Claude Sonnet 4.6Anthropic
1522.2
22.4
L—C—G—A—V—L—
L #32
⚔️ VS💰Plans🚀Try
#46
GPT-5.6 TerraOpenAI
1520.1
24.1
L #33
C #16
G—A—V—
L #14
L #33
⚔️ VS💰Plans🚀Try
#47
GPT-5.6 LunaOpenAI
1519.2
242.3
L—
C #19
G—A—V—L—
L #45
⚔️ VS💰Plans🚀Try
#48
Qwen3.7-MaxQwen
1515.1
203.8
L #41
C—G—A—V—L—
L #30
⚔️ VS💰Plans🚀Try
#49
GPT-5.5OpenAI
1512.1
18
L—C—G—A—V—L—
L #21
⚔️ VS💰Plans🚀Try
#50
Kimi K2.6Moonshot AI
1509.3
81.8
L—C—G—A—V—L—
L #37
⚔️ VS💰Plans🚀Try
#51
GLM-5.1Zhipu AI
1508.1
61.2
L—C—G—A—V—L—
L #35
⚔️ VS💰Plans🚀Try
#52
Gemini 3.5 FlashGoogle
1506
39.6
L—C—G—A—V—L—
L #23
⚔️ VS💰Plans🚀Try
#53
MiniMax-M3MiniMax
1484.1
163.4
L #43
C—G—
A #31
V—L—
L #55
⚔️ VS💰Plans🚀Try
#54
Qwen3.6-PlusQwen
1482.3
243.6
L—C—G—A—V—L—
L #39
⚔️ VS💰Plans🚀Try
#55
Kimi K2.7 CodeMoonshot AI
1474.2
45.1
L—C—G—A—V—L—L—
⚔️ VS💰Plans🚀Try
#56
GPT-5.4OpenAI
1461.7
22.7
L—C—G—A—V—L—
L #25
⚔️ VS💰Plans🚀Try
#57
Gemini 3.1 ProGoogle
1446.4
48.4
L #37
C—G—A—V—L—
L #14
⚔️ VS💰Plans🚀Try
#58
Gemini 3 ProGoogle
1440.7
48.4
L—C—G—A—V—L—
L #15
⚔️ VS💰Plans🚀Try
#59
GLM-4-PlusZhipu AI
1434.7
85.3
L—C—G—A—V—L—
L #52
⚔️ VS💰Plans🚀Try
#60
Gemini 3.5 Flash-LiteGoogle
1432
107.9
L—C—G—
A #32
V—L—
L #44
⚔️ VS💰Plans🚀Try
#61
GPT-5.4 miniOpenAI
1397.7
68.7
L—C—G—A—V—L—
L #46
⚔️ VS💰Plans🚀Try
#62
MiniMax-M2.7MiniMax
1397.7
211.4
L—C—G—A—V—L—
L #63
⚔️ VS💰Plans🚀Try
#63
DeepSeek-V3DeepSeek
1362.5
619.2
L—C—G—A—V—L—
L #59
⚔️ VS💰Plans🚀Try
#64
Grok 4.3xAI
1357.3
70.6
L—C—G—A—V—L—
L #51
⚔️ VS💰Plans🚀Try
#65
Claude Haiku 4.5Anthropic
1330.4
51.4
L—C—G—A—V—L—
L #65
⚔️ VS💰Plans🚀Try
#66
Mistral Medium 3.5Mistral AI
1263.1
74.2
L #44
C—G—
A #33
V—L—
L #58
⚔️ VS💰Plans🚀Try
#67
Qwen3.8-FlashQwen
1243.1
506
L—C—G—A—V—L—
L #70
⚔️ VS💰Plans🚀Try
#68
Grok 4.1 FastxAI
1241.6
370.8
L—C—G—A—V—L—
L #28
⚔️ VS💰Plans🚀Try
#69
Mistral Large 3Mistral AI
1232.3
30.5
L—C—G—A—V—L—
L #64
⚔️ VS💰Plans🚀Try
#70
Gemini 2.5 ProGoogle
1227.7
25.1
L—C—G—
A #28
V—L—
L #47
⚔️ VS💰Plans🚀Try
Rank color:#1#2#3#4-5#6-10#11+

The "Cross-board ranks" column shows this model's rank on the other boards at a glance (grey = not tested there). Boards use different methodologies, so raw values are not directly comparable.

LMArena Code board: LMSYS Coding Arena Elo ratings based on real developer blind tests on challenging code prompts. Sourced directly from arena.ai/leaderboard/code and mirrored 1:1.

ADSponsored
📊 8 benchmark boards
LMArena Code ↗ Coding

LMArena Code board: LMSYS Coding Arena Elo ratings based on real developer blind tests on challenging code prompts. Sourced directly from arena.ai/leaderboard/code and mirrored 1:1.

Coding Elo70 models
LMArena Agent ↗ Agent

LMArena Agent board: real agentic coding sessions scored by human blind votes (signal-based agent score). Sourced from arena.ai/leaderboard/agent and mirrored 1:1.

Agent Score45 models
CursorBench ↗ Agent

CursorBench is Cursor's first-party agentic coding benchmark (v4.0). Tasks come from real Cursor sessions with ambiguous, multi-file requirements. Metric: pass rate (%). Mirrored from Cursor's official cursor.com/cursorbench snapshot.

Pass rate%21 models
GDPx (SemiAnalysis) ↗ Agent

SemiAnalysis GDPx: General agentic benchmark evaluating 206 multi-hour tasks across 11 economically valuable domains (Chip Design, Cybersecurity, Biology, Finance, Legal, Quant Trading, Recursive Self-Improvement, Hardware Design, Materials Science, Alignment, and Software Engineering). Metric: Pass@1 (%). Sourced from gdpx.semianalysis.com.

Pass@1%20 models
Artificial Analysis ↗ Overall

Artificial Analysis Intelligence Index: Tracks reasoning, quality, and coding intelligence across 600+ models, continuously updated as new frontier models release.

Intelligence Index34 models
Vals AI ↗ Overall

Vals AI (Vals Index): Independent benchmarks evaluating frontier AI models on economically valuable real-world tasks across software, finance, cybersecurity, and agent autonomy. Metric: Accuracy (%). Sourced directly from vals.ai and mirrored 1:1.

Accuracy%28 models
LiveBench ↗ Coding

LiveBench Coding: Monthly refreshed contamination-resistant suite extracting code completion, generation, JavaScript, Python, and TypeScript performance.

Score%26 models
LMArena Text ↗ Chat

LMArena Text board: The global standard LMSYS Chatbot Arena tracking cross-domain text generation and complex instruction-following Elo ratings from arena.ai/leaderboard/text.

Arena Elo99 models

All rankings and numbers come directly from official sources; we only do name matching and aggregate presentation, no weighting or re-ranking. Boards use different methodologies, so do not compare raw values across boards.