Benchmark profile

Vals SAGE (SAGE)

Student Assessment with Generative Evaluation

Data verified July 17, 2026

How BenchLM shows SAGE

BenchLM mirrors the public Vals AI SAGE leaderboard captured from https://www.vals.ai/benchmarks/sage and updated by Vals on July 17, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

SAGE 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.

68 Vals rows5 task viewsprivate datasetTasks: Overall, Partial Differential Equations, Linear Algebra, Automata, CalculusDisplay only

SAGE on Vals AI Vals methodology Vals home

SAGE score on SAGE — July 17, 2026

BenchLM mirrors the published sage score view for SAGE. Claude Opus 4.7 leads the public snapshot at 56.10% , followed by Gemma 4 31b It (55.03%) and Claude Opus 4.8 (54.79%). BenchLM does not use these results to rank models overall.

Claude Opus 4.7

Anthropic

anthropic/claude-opus-4-7

56.10%

Overall —

Gemma 4 31b It

Google

google/gemma-4-31b-it

55.03%

Overall —

Claude Opus 4.8

Anthropic

anthropic/claude-opus-4-8

54.79%

Overall —

68 modelsExternal benchmark mirrorsCurrentDisplay onlyUpdated July 17, 2026

SAGE score table (68 models)

Score

Claude Opus 4.7Anthropic

56.10%

Gemma 4 31b ItGoogle

55.03%

Claude Opus 4.8Anthropic

54.79%

Kimi K3Moonshot AI

54.26%

GPT-5.6 SolOpenAI

52.56%

Claude Opus 4.5 20251101 ThinkingAnthropic

52.09%

Claude Fable 5Anthropic

51.89%

Gemini 3 Flash PreviewGoogle

51.85%

Claude Opus 4.6 ThinkingAnthropic

51.58%

GPT-5.5OpenAI

51.53%

GPT-5.4 MiniOpenAI

50.81%

MiniMax M3MiniMax

50.57%

Kimi K2.6Moonshot AI

50.22%

Gemini 3.5 FlashGoogle

49.88%

Kimi K2.5 ThinkingMoonshot AI

49.87%

Gemini 3.1 Flash Lite PreviewGoogle

49.54%

GPT-5.2OpenAI

49.27%

Claude Sonnet 5Anthropic

48.92%

Gemini 3.1 Pro PreviewGoogle

48.68%

Gemini 3 Pro PreviewGoogle

47.62%

GPT-5.6 TerraOpenAI

47.00%

Claude Sonnet 4.6Anthropic

46.58%

Muse Spark 1 1Meta

45.58%

Qwen3.6 27bAlibaba

45.58%

Claude Opus 4.5Anthropic

45.00%

Qwen3.6 PlusAlibaba

44.86%

Gemini 2.5 FlashGoogle

44.76%

GPT-5.6 LunaOpenAI

44.22%

Gemini 2.5 Flash Preview 09 2025Google

44.11%

GPT-5OpenAI

43.68%

GPT-5.4OpenAI

43.31%

Mimo V2.5Xiaomi

43.27%

GPT-5.1OpenAI

43.23%

GPT-5 MiniOpenAI

42.99%

Qwen3.5 FlashAlibaba

42.49%

Qwen3 VL PlusAlibaba

42.10%

Gemini 2.5 ProGoogle

41.92%

O3OpenAI

41.77%

O4 MiniOpenAI

41.06%

Gemini 2.5 Flash ThinkingGoogle

39.60%

Gemini 2.5 Flash Preview 09 2025 ThinkingGoogle

39.38%

Qwen3.7 PlusAlibaba

39.25%

Grok 4.20 0309 ReasoningSpaceXAI

38.24%

GPT-5.4 NanoOpenAI

38.08%

Mistral Medium 3.5Mistral AI

37.61%

InklingThinkingmachines

36.55%

Claude Sonnet 4.5 20250929 ThinkingAnthropic

36.06%

Llama4 Maverick Instruct BasicFireworks AI

35.70%

Claude Sonnet 4Anthropic

35.00%

Grok 4.5SpaceXAI

34.97%

Meta Llama Llama 4 Scout 17B 16E InstructTogether AI

34.83%

Claude Sonnet 4.5Anthropic

32.88%

Claude Sonnet 4 20250514 ThinkingAnthropic

32.01%

Claude Haiku 4.5 20251001 ThinkingAnthropic

31.82%

Claude Opus 4.1Anthropic

31.38%

Grok 4.1 Fast ReasoningSpaceXAI

31.33%

Gemini 2.5 Flash Lite Preview 09 2025 ThinkingGoogle

30.78%

Qwen3.5 Plus ThinkingAlibaba

30.40%

Claude Opus 4.1 20250805 ThinkingAnthropic

30.39%

GPT-5 NanoOpenAI

30.38%

Grok 4 Fast ReasoningSpaceXAI

29.76%

Gemini 2.5 Flash Lite Preview 09 2025Google

27.97%

Gemini 2.5 Flash LiteGoogle

27.53%

Grok 4 0709SpaceXAI

25.10%

Mistral Large 2512Mistral AI

24.59%

Grok 4.3SpaceXAI

19.74%

Grok 4 Fast Non ReasoningSpaceXAI

15.99%

Grok 4.1 Fast Non ReasoningSpaceXAI

11.91%

The published SAGE snapshot places Claude Opus 4.7 first at 56.10%. The third row is 1.32 points behind. The broader top-10 range is 4.57 points, so many of the published results sit in a relatively narrow band.

68 models have been evaluated on SAGE. 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. SAGE is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About SAGE

Year

2026

Tasks

Student assessment with generative evaluation

Format

Accuracy score

Difficulty

Education assessment reasoning

BenchLM mirrors the public Vals AI SAGE leaderboard as display-only external evidence. The captured snapshot preserves overall scores, task-level scores where Vals publishes them, uncertainty, latency, and cost-per-test metadata. It is excluded from BenchLM weighted rankings.

SAGE Public benchmark source

BenchLM freshness & provenance

Version

SAGE 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 SAGE measure?

Student Assessment with Generative Evaluation

Which model leads the published SAGE snapshot?

Claude Opus 4.7 currently leads the published SAGE snapshot with 56.10% sage score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on SAGE?

68 AI models are included in BenchLM's mirrored SAGE snapshot, based on the public leaderboard captured on July 17, 2026.

Last updated: July 17, 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.