Skip to main content
BenchLM

Scale Labs MultiNRC (MultiNRC)

We show this table for reference; we do not rank on it.

A Scale Labs public leaderboard mirrored as display-only reference data. It does not affect BenchLM rankings.

Scale score on MultiNRC — September 29, 2026 snapshot

We mirror the published scale score view for MultiNRC. Muse Spark 1.1 leads the public snapshot at 65.6%, followed by gpt-5-pro-2025-10-06 (65.2%) and gemini-3.1-pro-preview (64.7%). We do not use these results to rank models overall.

1Proprietary

Muse Spark 1.1

Meta

65.6%
Overall 65.77Context 1M
2

gpt-5-pro-2025-10-06

OpenAI

65.2%
3

gemini-3.1-pro-preview

Google

64.7%

44 modelsKnowledgeCurrentDisplay onlyUpdated September 29, 2026 snapshot

Scale score table (44 models)

Score
1
Muse Spark 1.1Meta · Closed
65.6%
2
gpt-5-pro-2025-10-06OpenAI
65.2%
3
gemini-3.1-pro-previewGoogle
64.7%
4
GPT-5.4 ProOpenAI · Closed
62.3%
5
Muse SparkMeta · Closed
59.0%
6
gemini-3-pro-previewGoogle
59.0%
7
gpt-5.4-2026-03-05 (xhigh thinking)OpenAI
58.3%
8
claude-opus-4-6-thinking-maxAnthropic
57.1%
9
gpt-5-2025-08-07OpenAI
52.1%
10
gpt-5.1-thinkingOpenAI
49%
11
o3-pro-2025-06-10-highOpenAI
49%
12
Claude Opus 4.5 ThinkingAnthropic · Closed
48.6%
13
claude-opus-4-6 (Non-Thinking)Anthropic
48.3%
14
o3-2025-04-16-highOpenAI
45.5%
15
Gemini-2.5-Pro-Preview-06-05Google
45.1%
16
o3-2025-04-16-mediumOpenAI
44.5%
17
GPT-5.2OpenAI · Closed
42.2%
18
Claude Opus 4.5Anthropic · Closed
41.2%
19
claude-opus-4-1-20250805-thinkingAnthropic
38.4%
20
Claude Sonnet 4.5 ThinkingAnthropic · Closed
35.8%
21
Kimi K2.5Moonshot AI · Open weight
35.2%
22
Claude-4-Opus-20250514-thinkingAnthropic
33.9%
23
claude-opus-4-1-20250805Anthropic
29.7%
24
Claude-4-Opus-20250514Anthropic
29%
25
Claude Sonnet 4.5Anthropic · Closed
28.1%
26
Claude-3.7-Sonnet-thinkingDeepseek
27.8%
27
Deepseek-R1-0528Deepseek
27.6%
28
Qwen3-235B-A22B-Thinking-2507Alibaba
27.1%
29
gemini-3.1-flash-lite-previewGoogle
25.0%
30
DeepSeek-R1DeepSeek · Open weight
24.3%
31
GPT-5 miniOpenAI · Closed
23.9%
32
DeepSeek V3.1DeepSeek · Open weight
23.6%
33
o4-mini (high)OpenAI · Closed
22.2%
34
GPT-4.1OpenAI · Closed
21.2%
35
kimi-k2-instructMoonshot AI
18.5%
36
Claude 4 SonnetAnthropic · Closed
18.4%
37
gpt-5.1-instantOpenAI
18.3%
38
Qwen3-235B-A22BAlibaba
17.6%
39
glm-4p5Z.AI
17.4%
40
GPT-OSS 120BOpenAI · Open weight
15.2%
41
GPT-4oOpenAI · Closed
12.4%
42
GPT-OSS 20BOpenAI · Open weight
10.4%
43
glm-4p5-airZ.AI
10.4%
44
Llama 4 MaverickMeta · Open weight
8.4%

How to read this leaderboard

Compare the published configurations as complete evaluation systems. The source can combine a base model, agent scaffold, tools, budget, and inference setting in each result.

Operator receipt: 44 sourced rows are currently displayable on this page; the leading published row is Muse Spark 1.1 at 65.6%.

Honest limit: This Scale table is display-only context, not benchmark provenance or a weighted model-only comparison.

MultiNRC on Scale Labs

How BenchLM shows MultiNRC

BenchLM mirrors 44 published rows from Scale Labs’ public MultiNRC leaderboard, captured on September 29, 2026 snapshot.

The table is display only. It is useful context for a published agent or model configuration, but it does not enter BenchLM’s overall or category rankings.

Snapshot

44 published rowsScale Labs sourceDisplay only
MultiNRC on Scale Labs

The published MultiNRC snapshot places Muse Spark 1.1 first at 65.6%. The third row is 0.9 points behind. The broader top-10 range is 16.6 points, so the table still separates the published systems.

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

About MultiNRC

Year

2026

Tasks

44 published rows

Format

Published Scale leaderboard score

Difficulty

External agent and model evaluation

BenchLM mirrors 44 published rows from the MultiNRC public table captured on September 29, 2026 snapshot.

Scale Labs leaderboardPublic benchmark source

Freshness and provenance

Version

MultiNRC 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.

Questions

What does MultiNRC measure?

A Scale Labs public leaderboard mirrored as display-only reference data. It does not affect BenchLM rankings.

Which model leads the published MultiNRC snapshot?

Muse Spark 1.1 currently leads the published MultiNRC snapshot with 65.6% scale score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on MultiNRC?

The September 29, 2026 snapshot snapshot contains 44 AI models.

Last updated: September 29, 2026 snapshot · mirrored from the public benchmark leaderboard

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.