GPT-6 Sol Max

GPT-6 Sol Max is a model from OpenAI. It has 6 scored results in 6 categories. It has no index.

  • Input: text
partialShows if the model has enough results for an index.
IndexOverall score out of 100.UnrankedMedian 50.7
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.—Median 62/s
First tokenSeconds to the first output token.N/AMedian 1.6 s
Input / 1MUS dollars per 1M input tokens.N/AMedian $0.6
Output / 1MUS dollars per 1M output tokens.N/AMedian $2.5
ContextMaximum tokens in one request.N/AMedian 262K
EloLMArena rating and rank.N/AMedian 1427

Capabilities

Score per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.1 counted65.0—
CodingCode writing and repair.1 counted72.4—
ReasoningLogic problems and puzzles.1 counted72.4—
MultimodalTasks with images and text.None countedN/A—
KnowledgeFacts and expert knowledge.1 counted71.5—
MultilingualTasks in many languages.None countedN/A—
InstructionTasks with strict rules in the prompt.1 counted66.3—
MathMath problems.1 counted75.3—

Results

6 counted
BenchmarkThe test name.ResultThe score from the source.PlacePlace among results on this benchmark.
LiveBench Data AnalysisReasoning81.2%5 / 62
LiveBench MathematicsMath96.4%6 / 62
LiveBench CodingCoding81.8%11 / 62
LiveBench LanguageKnowledge85.3%13 / 62
LiveBench ReasoningReasoning88.7%19 / 62
LiveBench Instruction FollowingInstruction68.6%32 / 62
LiveBench Agentic CodingAgentic52.9%34 / 62

Sources

BenchLM benchmark aggregationCC BY-NC 4.0 · Data from BenchLM.aiLiveBench, collected directlyNo licence stated for the leaderboard. The site repo serving these CSVs has no LICENSE; the harness repo carries upstream Apache-2.0 and MIT copies that cover code, not results

More from OpenAI

All models
GPT-6 Astra82.1GPT-6.1 Sol79.9GPT-5.5 Pro77.3GPT-6 Sol76.3GPT-5.4 Pro75.8