Skip to main content
benchlm.ai

Vals-hosted LiveCodeBench mirror (Vals LiveCodeBench mirror)

Vals AI implementation of LiveCodeBench with easy, medium, and hard task splits.

How BenchLM shows Vals LiveCodeBench mirror

BenchLM mirrors the public Vals AI Vals LiveCodeBench mirror leaderboard captured from https://www.vals.ai/benchmarks/lcb and updated by Vals on May 19, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

Vals LiveCodeBench mirror is display only on BenchLM. Vals proprietary or Vals-hosted aggregate views are useful context, but BenchLM does not use them as weighted ranking inputs or as a replacement for benchmark-native source records.

110 Vals rows4 task viewspublic datasetTasks: Overall, Easy, Medium, HardDisplay only
Vals LiveCodeBench mirror on Vals AIVals methodologyVals home

Vals LiveCodeBench score on Vals LiveCodeBench mirror — May 19, 2026

BenchLM mirrors the published vals livecodebench score view for Vals LiveCodeBench mirror. Gemini 3.1 Pro Preview leads the public snapshot at 88.48% , followed by GPT-5.2 Codex (87.99%) and Gemini 3.5 Flash (87.60%). BenchLM does not use these results to rank models overall.

1

Gemini 3.1 Pro Preview

Google

google/gemini-3.1-pro-preview

88.48%
Overall —
2

GPT-5.2 Codex

OpenAI

openai/gpt-5.2-codex

87.99%
Overall —
3

Gemini 3.5 Flash

Google

google/gemini-3.5-flash

87.60%
Overall —
110 modelsExternal benchmark mirrorsCurrentDisplay onlyUpdated May 19, 2026

The published Vals LiveCodeBench mirror snapshot is tightly clustered at the top: Gemini 3.1 Pro Preview sits at 88.48%, while the third row is only 0.88 points behind. The broader top-10 spread is 2.53 points, so many of the published scores sit in a relatively narrow band.

110 models have been evaluated on Vals LiveCodeBench mirror. The benchmark falls in the External benchmark mirrors category. BenchLM tracks this category separately from its weighted global scoring system, so these results are best compared on the dedicated Korean benchmark views. Vals LiveCodeBench mirror is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About Vals LiveCodeBench mirror

Year

2026

Tasks

Coding problem difficulty splits

Format

Accuracy score

Difficulty

Contamination-resistant coding problems

BenchLM keeps this separate from its canonical LiveCodeBench rows because it is a Vals-hosted implementation snapshot.

Vals LiveCodeBenchPublic benchmark source

BenchLM freshness & provenance

Version

Vals LiveCodeBench mirror 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Vals LiveCodeBench score table (110 models)

1
Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview
88.48%
2
GPT-5.2 Codexopenai/gpt-5.2-codex
87.99%
3
Gemini 3.5 Flashgoogle/gemini-3.5-flash
87.60%
4
DeepSeek V4 Prodeepseek/deepseek-v4-pro
87.48%
5
GPT-5.3 Codexopenai/gpt-5.3-codex
87.31%
6
Kimi K2.6 Thinkingkimi/kimi-k2.6-thinking
86.77%
7
GPT-5 Miniopenai/gpt-5-mini-2025-08-07
86.61%
8
GPT-5.1openai/gpt-5.1-2025-11-13
86.49%
9
Gemini 3 Pro Previewgoogle/gemini-3-pro-preview
86.41%
10
Qwen3.6 Plusalibaba/qwen3.6-plus
85.95%
11
GPT-5openai/gpt-5-2025-08-07
85.91%
12
Gemini 3 Flash Previewgoogle/gemini-3-flash-preview
85.59%
13
GPT-5.1 Codexopenai/gpt-5.1-codex
85.55%
14
GPT-5.2openai/gpt-5.2-2025-12-11
85.36%
15
Qwen3.5 Plus Thinkingalibaba/qwen3.5-plus-thinking
85.33%
16
GPT-5.5openai/gpt-5.5
85.30%
17
Claude Opus 4.7anthropic/claude-opus-4-7
85.07%
18
GPT-5 Codexopenai/gpt-5-codex
84.72%
19
Claude Opus 4.6 Thinkinganthropic/claude-opus-4-6-thinking
84.68%
20
Grok 4.3grok/grok-4.3
84.49%
21
Grok 4.20 0309 Reasoninggrok/grok-4.20-0309-reasoning
84.27%
22
GPT-5.4openai/gpt-5.4-2026-03-05
84.14%
23
GPT-5.4 Nanoopenai/gpt-5.4-nano-2026-03-17
84.01%
24
O3openai/o3-2025-04-16
83.91%
25
Kimi K2.5 Thinkingkimi/kimi-k2.5-thinking
83.87%
26
Claude Opus 4.5 20251101 Thinkinganthropic/claude-opus-4-5-20251101-thinking
83.67%
27
GPT-5.1 Codex Maxopenai/gpt-5.1-codex-max
83.56%
28
Qwen3.5 Flashalibaba/qwen3.5-flash
83.28%
29
Grok 4 0709grok/grok-4-0709
83.25%
30
GPT Oss 120bfireworks/gpt-oss-120b
83.23%
31
GLM 4.7zai/glm-4.7
82.23%
32
O4 Miniopenai/o4-mini-2025-04-16
82.21%
33
Claude Sonnet 4.6anthropic/claude-sonnet-4-6
82.09%
34
GLM 5 Thinkingzai/glm-5-thinking
81.87%
35
MiniMax M2.1minimax/MiniMax-M2.1
81.76%
36
GPT-5.4 Miniopenai/gpt-5.4-mini-2026-03-17
81.47%
37
GLM 5.1 Thinkingzai/glm-5.1-thinking
81.38%
38
GLM 4.6zai/glm-4.6
81.04%
39
DeepSeek V3p2 Thinkingfireworks/deepseek-v3p2-thinking
80.69%
40
Grok 4.1 Fast Reasoninggrok/grok-4-1-fast-reasoning
80.64%
41
GPT Oss 20bfireworks/gpt-oss-20b
80.39%
42
Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview
80.12%
43
MiniMax M2.7minimax/MiniMax-M2.7
79.93%
44
MiniMax M2.5 Lightningminimax/MiniMax-M2.5-Lightning
79.21%
45
Gemini 2.5 Pro Preview 03 25google/gemini-2.5-pro-preview-03-25
79.16%
46
Grok 4 Fast Reasoninggrok/grok-4-fast-reasoning
78.97%
47
Qwen3 Maxalibaba/qwen3-max
78.22%
48
Grok 3 Mini Fast High Reasoninggrok/grok-3-mini-fast-high-reasoning
76.22%
49
Gemini 2.5 Flash Preview 09 2025 Thinkinggoogle/gemini-2.5-flash-preview-09-2025-thinking
76.21%
50
Gemini 2.5 Flash Preview 09 2025google/gemini-2.5-flash-preview-09-2025
75.06%
51
Claude Opus 4.5anthropic/claude-opus-4-5-20251101
75.03%
52
Magistral Medium 2509mistralai/magistral-medium-2509
74.86%
53
Claude Sonnet 4.5 20250929 Thinkinganthropic/claude-sonnet-4-5-20250929-thinking
73.00%
54
Magistral Small 2509mistralai/magistral-small-2509
72.13%
55
O3 Miniopenai/o3-mini-2025-01-31
71.48%
56
Gemini 2.5 Flash Lite Preview 09 2025 Thinkinggoogle/gemini-2.5-flash-lite-preview-09-2025-thinking
71.39%
57
Qwen3 235b A22bfireworks/qwen3-235b-a22b
70.62%
58
Moonshotai Kimi K2 Instructtogether/moonshotai/Kimi-K2-Instruct
70.45%
59
DeepSeek R1fireworks/deepseek-r1
70.22%
60
GPT-5 Nanoopenai/gpt-5-nano-2025-08-07
70.22%
61
Claude Opus 4 20250514 Thinkinganthropic/claude-opus-4-20250514-thinking
70.19%
62
DeepSeek V3p2fireworks/deepseek-v3p2
69.86%
63
Gemini 2.5 Flash Lite Preview 09 2025google/gemini-2.5-flash-lite-preview-09-2025
67.67%
64
GLM 4.5zai/glm-4.5
67.45%
65
Qwen3 Max Previewalibaba/qwen3-max-preview
66.91%
66
Claude Opus 4.1 20250805 Thinkinganthropic/claude-opus-4-1-20250805-thinking
66.46%
67
Grok 3 Mini Fast Low Reasoninggrok/grok-3-mini-fast-low-reasoning
66.27%
68
DeepSeek V3 0324fireworks/deepseek-v3-0324
65.48%
69
Claude Opus 4.1anthropic/claude-opus-4-1-20250805
64.56%
70
Kimi K2 Thinkingkimi/kimi-k2-thinking
63.15%
71
Claude Opus 4anthropic/claude-opus-4-20250514
62.63%
72
Claude Sonnet 4 20250514 Thinkinganthropic/claude-sonnet-4-20250514-thinking
62.39%
73
Grok Code Fast 1grok/grok-code-fast-1
61.97%
74
Claude 3.7 Sonnet 20250219 Thinkinganthropic/claude-3-7-sonnet-20250219-thinking
60.44%
75
Claude Sonnet 4anthropic/claude-sonnet-4-20250514
59.67%
76
Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561 Thinkingtogether/langston/nim/nvidia/llama-3.3-nemotron-super-49b-v1-42e84561-thinking
58.37%
77
GPT-4.1 Miniopenai/gpt-4.1-mini-2025-04-14
58.16%
78
Gemini 2.5 Flash Preview 04 17google/gemini-2.5-flash-preview-04-17
56.94%
79
Claude 3.7 Sonnetanthropic/claude-3-7-sonnet-20250219
56.66%
80
Mistral Large 2512mistralai/mistral-large-2512
55.34%
81
GPT-4.1openai/gpt-4.1-2025-04-14
54.67%
82
Grok 3grok/grok-3
52.90%
83
Devstral 2512mistralai/devstral-2512
51.84%
84
O1openai/o1-2024-12-17
50.26%
85
Claude 3.5 Sonnetanthropic/claude-3-5-sonnet-20241022
49.63%
86
Llama4 Maverick Instruct Basicfireworks/llama4-maverick-instruct-basic
47.25%
87
Gemini 2.5 Flash Preview 04 17 Thinkinggoogle/gemini-2.5-flash-preview-04-17-thinking
46.87%
88
Grok 4 Fast Non Reasoninggrok/grok-4-fast-non-reasoning
46.09%
89
Mistral Medium 2505mistralai/mistral-medium-2505
44.84%
90
Gemini 2.0 Flash 001google/gemini-2.0-flash-001
43.61%
91
GPT-4oopenai/gpt-4o-2024-11-20
43.44%
92
Labs Devstral Small 2512mistralai/labs-devstral-small-2512
43.18%
93
GPT-4.1 Nanoopenai/gpt-4.1-nano-2025-04-14
42.72%
94
Grok 4.1 Fast Non Reasoninggrok/grok-4-1-fast-non-reasoning
42.62%
95
Claude 3.5 Haikuanthropic/claude-3-5-haiku-20241022
41.92%
96
Gemini 1.5 Pro 002google/gemini-1.5-pro-002
41.72%
97
Claude Haiku 4.5 20251001 Thinkinganthropic/claude-haiku-4-5-20251001-thinking
41.17%
98
Grok 2 1212grok/grok-2-1212
38.68%
99
Meta Llama Llama 4 Scout 17B 16E Instructtogether/meta-llama/Llama-4-Scout-17B-16E-Instruct
38.54%
100
Mistral Large 2411mistralai/mistral-large-2411
37.09%
101
Gemini 1.5 Flash 002google/gemini-1.5-flash-002
36.91%
102
Meta Llama Llama 3.3 70B Instruct Turbotogether/meta-llama/Llama-3.3-70B-Instruct-Turbo
36.34%
103
Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561together/langston/nim/nvidia/llama-3.3-nemotron-super-49b-v1-42e84561
36.31%
104
Command A 03 2025cohere/command-a-03-2025
35.07%
105
Mistral Small 2503mistralai/mistral-small-2503
31.82%
106
GPT-4o Miniopenai/gpt-4o-mini-2024-07-18
26.42%
107
Jamba Large 1.6ai21labs/jamba-large-1.6
22.32%
108
Command R Pluscohere/command-r-plus
18.24%
109
Mistral Small 2402mistralai/mistral-small-2402
15.78%
110
Jamba Mini 1.6ai21labs/jamba-mini-1.6
9.92%

FAQ

What does Vals LiveCodeBench mirror measure?

Vals AI implementation of LiveCodeBench with easy, medium, and hard task splits.

Which model leads the published Vals LiveCodeBench mirror snapshot?

Gemini 3.1 Pro Preview currently leads the published Vals LiveCodeBench mirror snapshot with 88.48% vals livecodebench score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on Vals LiveCodeBench mirror?

110 AI models are included in BenchLM's mirrored Vals LiveCodeBench mirror snapshot, based on the public leaderboard captured on May 19, 2026.

Last updated: May 19, 2026 · mirrored from the public benchmark leaderboard

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.