Skip to main content

Benchmark profile

Vals CorpFin v2 (CorpFin v2)

Vals AI private benchmark for understanding long-context credit agreements.

Data verified

How BenchLM shows CorpFin v2

BenchLM mirrors the public Vals AI CorpFin v2 leaderboard captured from https://www.vals.ai/benchmarks/corp_fin_v2 and updated by Vals on July 16, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

CorpFin v2 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.

123 Vals rows4 task viewsprivate datasetTasks: Overall, Exact Pages, Max Fitting Context, Shared Max ContextDisplay only

CorpFin v2 score on CorpFin v2 — July 16, 2026

BenchLM mirrors the published corpfin v2 score view for CorpFin v2. Claude Fable 5 leads the public snapshot at 71.83% , followed by Kimi K3 (71.56%) and Muse Spark 1 1 (71.29%). BenchLM does not use these results to rank models overall.

123 modelsExternal benchmark mirrorsCurrentDisplay onlyUpdated July 16, 2026

CorpFin v2 score table (123 models)

Score
1
71.83%
2
Kimi K3Moonshot AI
71.56%
3
71.29%
4
InklingThinkingmachines
68.57%
5
Grok 4.3SpaceXAI
68.53%
6
GPT-5.5OpenAI
68.42%
7
68.26%
8
MiniMax M3MiniMax
68.10%
9
Qwen3 MaxAlibaba
68.03%
10
Grok 4.5SpaceXAI
67.41%
11
67.02%
12
66.98%
13
66.90%
14
Kimi K2.6Moonshot AI
66.74%
15
66.71%
16
66.47%
18
GLM 5.2Zhipu AI
66.12%
19
66.08%
20
Grok 4 0709SpaceXAI
66.05%
21
65.97%
22
GPT-5.2OpenAI
65.89%
24
65.31%
25
65.31%
26
65.31%
27
GPT-5.4OpenAI
65.27%
28
65.11%
30
64.69%
32
GLM 5.1Zhipu AI
64.45%
33
64.38%
34
64.22%
35
GPT-5.1OpenAI
63.83%
36
63.71%
37
63.67%
38
63.67%
39
63.56%
40
GPT-4.1OpenAI
63.05%
41
62.90%
42
62.31%
44
61.93%
45
61.42%
46
61.38%
47
61.30%
49
61.19%
50
61.19%
52
GPT-5OpenAI
61.07%
53
61.03%
54
GLM 4.5Zhipu AI
60.96%
55
60.92%
56
60.80%
57
60.80%
59
Kimi K2 ThinkingMoonshot AI
60.57%
61
60.30%
62
60.18%
63
Mimo V2.5Xiaomi
59.91%
66
O3OpenAI
59.71%
67
Grok 3SpaceXAI
59.71%
68
59.60%
71
O4 MiniOpenAI
58.97%
73
58.90%
74
58.78%
76
GPT Oss 120bFireworks AI
58.24%
77
Laguna M.1Poolside
58.16%
78
57.93%
80
GLM 4.6Zhipu AI
56.84%
81
Laguna Xs.2Poolside
56.33%
83
Qwen3 MaxAlibaba
55.94%
84
DeepSeek V3 0324Fireworks AI
54.74%
85
54.70%
86
54.66%
88
DeepSeek R1Fireworks AI
54.12%
89
53.61%
90
GPT Oss 20bFireworks AI
53.15%
91
52.95%
93
DeepSeek V3Fireworks AI
52.49%
94
DeepSeek V3p1Fireworks AI
51.48%
95
Grok 2 1212SpaceXAI
51.13%
96
50.97%
97
50.82%
98
50.74%
99
50.39%
100
49.73%
101
DeepSeek V3p2Fireworks AI
47.94%
102
47.40%
104
GLM 4.7Zhipu AI
46.39%
105
45.96%
106
GPT-4oOpenAI
45.92%
107
45.45%
108
O3 MiniOpenAI
45.30%
109
44.17%
110
44.02%
112
42.08%
113
41.53%
114
40.52%
115
GPT-4oOpenAI
39.43%
116
39.43%
118
38.19%
119
38.03%
121
33.88%
122
33.72%
123
28.63%

The published CorpFin v2 snapshot places Claude Fable 5 first at 71.83%. The third row is 0.54 points behind. The broader top-10 range is 4.43 points, so many of the published results sit in a relatively narrow band.

123 models have been evaluated on CorpFin v2. The benchmark falls in the External benchmark mirrors category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. CorpFin v2 is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About CorpFin v2

Year

2026

Tasks

Credit-agreement understanding tasks

Format

Accuracy score

Difficulty

Professional finance document reasoning

The Vals CorpFin v2 page reports overall, exact-page, max-fitting-context, and shared-max-context task views. BenchLM keeps it display only.

BenchLM freshness & provenance

Version

CorpFin v2 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.

FAQ

What does CorpFin v2 measure?

Vals AI private benchmark for understanding long-context credit agreements.

Which model leads the published CorpFin v2 snapshot?

Claude Fable 5 currently leads the published CorpFin v2 snapshot with 71.83% corpfin v2 score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on CorpFin v2?

123 AI models are included in BenchLM's mirrored CorpFin v2 snapshot, based on the public leaderboard captured on July 16, 2026.

Last updated: July 16, 2026 · mirrored from the public benchmark leaderboard

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.