Vals Multimodal Index v1.2 (Vals Multimodal Index)
We show this table for reference; we do not rank on it.
Vals AI multimodal composite across finance, coding, education, and mortgage-tax task families.
Evaluation results published by Vals AI. Published leaderboard aggregates and evaluation settings only. Dataset contents are not included.
Vals Multimodal Index score on Vals Multimodal Index — August 11, 2026
We mirror the published vals multimodal index score view for Vals Multimodal Index. Claude Fable 5 leads the public snapshot at 74.15%, followed by Claude Opus 5 (73.90%) and Kimi K3 (73.42%). We do not use these results to rank models overall.
Claude Fable 5
Anthropic
Claude Opus 5
Anthropic
Kimi K3
Moonshot AI
33 modelsMultimodal & GroundedCurrentDisplay onlyUpdated August 11, 2026
Vals Multimodal Index score table (33 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Claude Fable 5Anthropic | 74.15% | Not reported | Closed |
| 2 | Claude Opus 5Anthropic | 73.90% | Not reported | Closed |
| 3 | Kimi K3Moonshot AI | 73.42% | Not reported | Open |
| 4 | GPT-5.6 SolOpenAImax reasoning | 72.64% | Not reported | Closed |
| 5 | Claude Opus 4.8Anthropic | 70.89% | Not reported | Closed |
| 6 | Muse Spark 1.2Meta | 69.80% | Not reported | Closed |
| 7 | GPT-5.6 LunaOpenAImax reasoning | 69.06% | Not reported | Closed |
| 8 | Claude Sonnet 5Anthropic | 68.83% | Not reported | Closed |
| 9 | GPT-5.5OpenAIxhigh reasoning | 68.07% | Not reported | Closed |
| 10 | Claude Opus 4.7Anthropic | 67.36% | Not reported | Closed |
| 11 | Muse Spark 1.1Metaxhigh reasoning | 66.74% | Not reported | Closed |
| 12 | Qwen3.8 MaxAlibaba | 65.39% | Not reported | Open |
| 13 | Gemini 3.6 FlashGooglehigh reasoning | 65.08% | Not reported | Closed |
| 14 | GPT-5.6 TerraOpenAIxhigh reasoning | 65.07% | Not reported | Closed |
| 15 | Grok 4.5xAIhigh reasoning | 63.42% | Not reported | Closed |
| 16 | Gemini 3.5 FlashGooglehigh reasoning | 62.93% | Not reported | Closed |
| 17 | Claude Sonnet 4.6Anthropic | 60.57% | Not reported | Closed |
| 18 | MiniMax M3MiniMax | 59.97% | Not reported | Open |
| 19 | Kimi K2.6Moonshot AI | 56.43% | Not reported | Open |
| 20 | Gemini 3.1 Pro PreviewGooglehigh reasoning | 56.07% | Not reported | Unknown |
| 21 | GPT-5.4 miniOpenAIxhigh reasoning | 54.22% | Not reported | Closed |
| 22 | Qwen3.7 PlusAlibaba | 53.89% | Not reported | Closed |
| 23 | MiMo-V2.5Xiaomi | 52.77% | Not reported | Closed |
| 24 | Gemini 3 Flash PreviewGooglehigh reasoning | 52.19% | Not reported | Unknown |
| 25 | Qwen3.6 PlusAlibaba | 51.52% | Not reported | Closed |
| 26 | Inkling-SmallThinking Machines Lab0.99 reasoning | 50.05% | Not reported | Open |
| 27 | InklingThinking Machines Lab0.99 reasoning | 49.40% | Not reported | Open |
| 28 | GPT-5.4 nanoOpenAIhigh reasoning | 47.64% | Not reported | Closed |
| 29 | Grok 4.3xAIhigh reasoning | 43.29% | Not reported | Closed |
| 30 | Claude Haiku 4.5 ThinkingAnthropic | 42.88% | Not reported | Closed |
| 31 | Gemini 3.1 Flash Lite PreviewGooglehigh reasoning | 41.35% | Not reported | Unknown |
| 32 | Grok 4.20 0309 ReasoningxAI | 39.06% | Not reported | Unknown |
| 33 | Mistral Medium 3.5Mistral AIhigh reasoning | 34.77% | Not reported | Unknown |
How Vals Multimodal Index is shown here
BenchLM mirrors the public Vals AI Vals Multimodal Index leaderboard captured from https://www.vals.ai/benchmarks/vals_multimodal_index and updated by Vals on August 11, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.
Vals Multimodal Index 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.
Snapshot
The published Vals Multimodal Index snapshot places Claude Fable 5 first at 74.15%. The third row is 0.73 points behind. The broader top-10 range is 6.79 points, so many of the published results sit in a relatively narrow band.
33 models have been evaluated on Vals Multimodal Index. The benchmark falls in the Multimodal & Grounded category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. Vals Multimodal Index is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Vals Multimodal Index
Year
2026
Tasks
Finance, coding, education, and mortgage-tax components
Format
Composite score
Difficulty
Private multimodal economic-work benchmark composite
BenchLM mirrors the Vals Multimodal Index as a display-only external composite with task-level component scores preserved in the Vals snapshot.
Freshness and provenance
Version
Vals Multimodal Index 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
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.
Questions
What does Vals Multimodal Index measure?
Vals AI multimodal composite across finance, coding, education, and mortgage-tax task families.
Which model leads the published Vals Multimodal Index snapshot?
Claude Fable 5 currently leads the published Vals Multimodal Index snapshot with 74.15% vals multimodal index score. BenchLM shows this benchmark for display only and does not use it in overall rankings.
How many models are evaluated on Vals Multimodal Index?
The August 11, 2026 snapshot contains 33 AI models.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.