Scale Labs MultiChallenge (Large Language Models) (MultiChallenge (Large Language Models))
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 MultiChallenge (Large Language Models) — November 12, 2026 snapshot
We mirror the published scale score view for MultiChallenge (Large Language Models). GPT-6.1 Sol leads the public snapshot at 82.0%, followed by Gemini 4 Argon (78.7%) and Muse Spark (75.5%). We do not use these results to rank models overall.
GPT-6.1 Sol
OpenAI
MultiChallenge (Large Language Models)
Gemini 4 Argon
MultiChallenge (Large Language Models)
Muse Spark
Meta
MultiChallenge (Large Language Models)
36 modelsInstruction FollowingCurrentDisplay onlyUpdated November 12, 2026 snapshot
Scale score table (36 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | GPT-6.1 SolOpenAIMultiChallenge (Large Language Models) | 82.0% | Not reported | Closed |
| 2 | Gemini 4 ArgonGoogleMultiChallenge (Large Language Models) | 78.7% | Not reported | Closed |
| 3 | Muse SparkMetaMultiChallenge (Large Language Models) | 75.5% | Not reported | Closed |
| 4 | Muse Spark 1.1\nMetaMultiChallenge (Large Language Models) | 75.3% | Not reported | Unknown |
| 5 | Deepseek V4p1 Flash 1017DeepseekMultiChallenge (Large Language Models) | 74.9% | Not reported | Unknown |
| 6 | GPT-6.1 TerraOpenAIMultiChallenge (Large Language Models) | 74.2% | Not reported | Closed |
| 7 | Muse Spark OpenMetaMultiChallenge (Large Language Models) | 74.2% | Not reported | Open |
| 8 | gemini-3.1-pro-previewGoogleMultiChallenge (Large Language Models) | 71.4% | Not reported | Unknown |
| 9 | GPT-5.4 ProOpenAIMultiChallenge (Large Language Models) | 69.2% | Not reported | Closed |
| 10 | Claude Haiku 5.5AnthropicMultiChallenge (Large Language Models) | 67.7% | Not reported | Closed |
| 11 | gemini-3-pro-previewGoogleMultiChallenge (Large Language Models) | 65.7% | Not reported | Unknown |
| 12 | gpt-5.1-2025-11-13-thinkingOpenAIMultiChallenge (Large Language Models) | 63.4% | Not reported | Unknown |
| 13 | gpt-5-thinkingOpenAIMultiChallenge (Large Language Models) | 63.2% | Not reported | Unknown |
| 14 | o3-pro-2025-06-10-reasoning-highOpenAIMultiChallenge (Large Language Models) | 62.4% | Not reported | Unknown |
| 15 | Kimi K2.5Moonshot AIMultiChallenge (Large Language Models) | 61.4% | Not reported | Open |
| 16 | gemini-3.1-flash-lite-previewGoogleMultiChallenge (Large Language Models) | 60.6% | Not reported | Unknown |
| 17 | gpt-5-mini-thinkingOpenAIMultiChallenge (Large Language Models) | 59.0% | Not reported | Unknown |
| 18 | Claude Opus 4.5 ThinkingAnthropicMultiChallenge (Large Language Models) | 59.0% | Not reported | Closed |
| 19 | Claude 4 Opus ThinkingAnthropicMultiChallenge (Large Language Models) | 58.6% | Not reported | Closed |
| 20 | claude-opus-4-1-20250805-thinkingAnthropicMultiChallenge (Large Language Models) | 57.2% | Not reported | Unknown |
| 21 | claude-4-sonnet-thinkingAnthropicMultiChallenge (Large Language Models) | 57.1% | Not reported | Unknown |
| 22 | o3-2025-04-16-reasoning-highOpenAIMultiChallenge (Large Language Models) | 56.6% | Not reported | Unknown |
| 23 | claude-opus-4-6 (Non-Thinking)AnthropicMultiChallenge (Large Language Models) | 56.0% | Not reported | Unknown |
| 24 | kimi-k2-thinkingMoonshot AIMultiChallenge (Large Language Models) | 55.4% | Not reported | Unknown |
| 25 | Claude Sonnet 4.5 ThinkingAnthropicMultiChallenge (Large Language Models) | 55.3% | Not reported | Closed |
| 26 | Gemini 2.5 ProGoogleMultiChallenge (Large Language Models) | 53.6% | Not reported | Closed |
| 27 | Claude 3.7 Sonnet ThinkingAnthropicMultiChallenge (Large Language Models) | 51.6% | Not reported | Closed |
| 28 | gpt-5.1-2025-11-13-instantOpenAIMultiChallenge (Large Language Models) | 51.2% | Not reported | Unknown |
| 29 | Claude Haiku 4.5 ThinkingAnthropicMultiChallenge (Large Language Models) | 50.5% | Not reported | Closed |
| 30 | deepseek-v3p1DeepseekMultiChallenge (Large Language Models) | 46.1% | Not reported | Unknown |
| 31 | GPT-OSS 120BOpenAIMultiChallenge (Large Language Models) | 45.3% | Not reported | Open |
| 32 | o4-mini-2025-04-16-reasoning-highOpenAIMultiChallenge (Large Language Models) | 44.9% | Not reported | Unknown |
| 33 | Qwen3 235B-A22BAlibabaMultiChallenge (Large Language Models) | 41.2% | Not reported | Open |
| 34 | GPT-4.1OpenAIMultiChallenge (Large Language Models) | 39.4% | Not reported | Closed |
| 35 | claude-opus-4-6-thinking-maxAnthropicMultiChallenge (Large Language Models) | 37.1% | Not reported | Unknown |
| 36 | Gemini 2.0 FlashGoogleMultiChallenge (Large Language Models) | 36.4% | Not reported | Closed |
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: 36 sourced rows are currently displayable on this page; the leading published row is GPT-6.1 Sol at 82.0%.
Honest limit: This Scale table is display-only context, not benchmark provenance or a weighted model-only comparison.
How BenchLM shows MultiChallenge (Large Language Models)
BenchLM mirrors 36 published rows from Scale Labs’ public MultiChallenge (Large Language Models) leaderboard, captured on November 12, 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
The published MultiChallenge (Large Language Models) snapshot places GPT-6.1 Sol first at 82.0%. The third row is 6.4 points behind. The broader top-10 range is 14.3 points, so the table still separates the published systems.
36 models have been evaluated on MultiChallenge (Large Language Models). The benchmark falls in the Instruction Following category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. MultiChallenge (Large Language Models) is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About MultiChallenge (Large Language Models)
Year
2026
Tasks
36 published rows
Format
Published Scale leaderboard score
Difficulty
External agent and model evaluation
BenchLM mirrors 36 published rows from the MultiChallenge (Large Language Models) public table captured on November 12, 2026 snapshot.
Freshness and provenance
Version
MultiChallenge (Large Language Models) 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 MultiChallenge (Large Language Models) measure?
A Scale Labs public leaderboard mirrored as display-only reference data. It does not affect BenchLM rankings.
Which model leads the published MultiChallenge (Large Language Models) snapshot?
GPT-6.1 Sol currently leads the published MultiChallenge (Large Language Models) snapshot with 82.0% scale score. BenchLM shows this benchmark for display only and does not use it in overall rankings.
How many models are evaluated on MultiChallenge (Large Language Models)?
The November 12, 2026 snapshot snapshot contains 36 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.