Inference
EN
OpenAI

GPT-6.1 Sol

GPT-6.1 Sol is a reasoning model from OpenAI in the GPT-6 family. 55 benchmarks count toward its index, in 8 categories.

  • Closed weights
  • Reasoning
  • Input: file · image · text
  • Released 2026-09
RankedEnough results for an index.
IndexAll categories, out of 100.78.0±2.2Median 51.0
CoverageIndex weight that has results.100%
SpeedOutput tokens per second.34/sMedian 67/s
First tokenSeconds to the first output token.4.6 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$2.00batch $1.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$10.00batch $5.00 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.1.05MMedian 262K
EloLMArena rating and rank.1446±6 #62Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
openai/gpt-6.1-sol
Max outputMaximum output tokens per request.
128K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.10
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
5 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
Amazon BedrockCloud$2.00$10.001.05M
DigitalOceanCloud$2.00$10.001.05M
Microsoft AzureCloud$2.00$10.001.05M
OpenAIMaker$2.00$10.001.05M
OpenRouterRouter$2.00$10.001.05M

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.17 counted76.08 / 218
CodingCode writing and repair.10 counted75.89 / 209
ReasoningLogic problems and puzzles.11 counted81.84 / 227
MultimodalTasks with images and text.3 counted72.27 / 84
KnowledgeFacts and expert knowledge.7 counted78.05 / 268
MultilingualTasks in many languages.1 counted73.2—
InstructionTasks with strict prompt rules.1 counted76.6—
MathMath problems.5 counted78.22 / 168

Results

55 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
OTIS Mock AIME 2024-2025Math100.0%1 / 200
EuroEval PortugueseMultilingual74.8% ±0.41 / 182
EuroEval ItalianMultilingual78.5% ±0.31 / 174
EuroEval SwedishMultilingual77.5% ±0.42 / 195
EuroEval DutchMultilingual74.0% ±0.42 / 175
FrontierMath-Tier-4-v2-PrivateMath100.0%1 / 72
EuroEval PolishMultilingual66.9% ±0.33 / 209
ARC-AGI-2 (semi-private)Reasoning94.2%2 / 131
EuroEval GermanMultilingual67.1% ±0.53 / 191
EuroEval FrenchMultilingual76.1% ±0.43 / 178
EuroEval SpanishMultilingual68.2% ±0.53 / 176
Critical Physics TasksReasoningAbove fitFar above this model's index.31.7%4 / 214
FrontierMath-Tiers-1-3-v2-PrivateMath93.7% ±1.42 / 90
GPQA diamond · Epoch AIKnowledge95.4% ±1.45 / 217
Artificial Analysis MMMU-ProMultimodal86.0%3 / 113
LiveBench LanguageKnowledge90.1%2 / 66
Artificial Analysis Omniscience AccuracyKnowledge61.8%5 / 164
LiveBench ReasoningReasoning92.6%3 / 80
LiveBench Data AnalysisReasoning82.7%3 / 80
Mystery Game PuzzlesReasoningAbove fitFar above this model's index.80.0% ±4.03 / 80
ExtractBench: schema-guided enterprise document extraction (LlamaIndex leaderboard)Multimodal95.3%1 / 26
MazeBench: visual spatial reasoning in a 3D open worldAgentic17.0%2 / 48
Chess PuzzlesReasoning68.0% ±4.74 / 90
Artificial Analysis Humanity's Last ExamKnowledge52.9%10 / 222
LiveBench MathematicsMath96.8%3 / 66
SimpleQA VerifiedKnowledge73.9% ±1.45 / 91
Code MigrationCoding65.1% ±4.45 / 77
IOICoding96.9% ±3.13 / 45
LiveBench Coding 2.0Coding96.0%4 / 52
LiveBench Long AgenticAgentic73.0%4 / 51
Furniture AssemblyReasoning80.0% ±4.93 / 37
Vibe Code Bench v1.1Coding88.9% ±1.910 / 123
Artificial Analysis Long Context ReasoningReasoning83.0%18 / 217
ARC-AGI-1 (semi-private)Reasoning96.5%11 / 126
ProofBench v1.1Math99.0% ±1.06 / 54
Hermes Index: agentic capability and cost per task inside Hermes AgentAgentic57.1%2 / 16
Terminal-Bench 4.1.0Agentic60.9% ±1.53 / 23
Terminal-Bench 4.0.0Agentic58.2% ±1.63 / 23
Artificial Analysis GDP.pdfAgentic31.0%3 / 22
Terminal-Bench 4.0Agentic55.1% ±1.89 / 61
Agent Arena steerabilityAgentic13.48 / 54
Agent Arena task outcomeAgentic11.78 / 54
ARC-AGI-3 (semi-private)Reasoning52.7%4 / 26
Bug Hunt BenchCoding44.3%4 / 25
LiveBench Browser UseAgentic88.0%8 / 49
GDPval-AA normalizedAgentic54.8%22 / 133
LiveBench Instruction FollowingInstruction74.2%12 / 66
Medical Long Context Reasoning (MLCR-AA)Reasoning38.3%4 / 22
LiveBench DesignAgentic82.1%10 / 52
LiveBench ResearchAgentic83.5%11 / 52
FrontierCode 2.0Coding39.0%6 / 27
EBR-benchReasoning54.3% ±5.88 / 31
ExploitGymAgentic35.1%6 / 23
LiveBench CodingCoding80.4%18 / 66
HealthBench ProfessionalKnowledge64.2%5 / 18
MirrorCodeCoding61.1% ±11.311 / 39
DeepSWEAgentic71.9%18 / 59
Artificial Analysis SciCodeCoding54.2%35 / 114
ParseBench: document parsing for AI agents (LlamaIndex leaderboard)Multimodal69.6%12 / 39
HealthBench HardKnowledge36.2%4 / 13
Artificial Analysis Terminal-Bench v4.0Agentic56.1%8 / 22
Multimodal Knowledge Agent BenchmarkAgentic47.7%7 / 19
Artificial Analysis AnalystAgentAgentic50.0%10 / 26
Agent Arena command recoveryAgentic4.520 / 50
Artificial Analysis AutomationBenchAgentic64.9%10 / 23
ScarfBench: Cross-Framework Application Migration in Enterprise JavaCoding68.4%5 / 11
LiveBench Agentic CodingAgentic54.5%31 / 66
Terminal-Bench-Science 0.1Agentic57.0%7 / 11
AutomationBenchAgentic36.1%22 / 32
ARC-AGI-2 (public eval)Reasoning97.1%6 / 122
ARC-AGI-1 (public eval)Reasoning98.5%6 / 116
SRE BenchOther50.8% ±3.12 / 36
FrontierMath-ErdosMath4.4% ±2.51 / 15
Terminal-Bench-ScienceScience52.9% ±6.03 / 44
SAGEEducation90.5% ±1.52 / 29
ProgramBenchCoding3.0% ±1.26 / 65
BioMysteryBenchScience79.6% ±0.73 / 30
EMBFinance70.8% ±2.714 / 75
MysteryMechanismScience46.4% ±3.46 / 30
MedScribeHealthcare86.5% ±1.922 / 109
SpireClimbOther23.8% ±3.72 / 9
MedCodeHealthcare48.8% ±2.128 / 106
SpireClimb (Slay the Spire 2, Ascension 10)Other29.3% ±5.33 / 9
SpireClimb (Slay the Spire 1, Ascension 20 Heart)Other18.3% ±5.03 / 9
Legal Research BenchLegal38.5% ±3.439 / 90
Finance Agent (v2)Finance52.0% ±0.337 / 79
Tax Agent BenchFinance21.8% ±3.049 / 82
Public Benefits Bench v1.1Public benefits59.3% ±1.335 / 50
HLAB-Vals-1.1Legal6.3% ±1.826 / 35
CyberBench v1.1Other39.3% ±2.746 / 48
Hermes Bench: 150 agent tasks graded on the files and state the agent leaves behindAgentic73.9%—
HealthBench Professional raw scoreKnowledge67.2%—
BioBasicsBench: everyday wet-lab scienceReasoning62.6%—
HealthBench length-adjusted scoreKnowledge58.5%—
HealthBench raw scoreKnowledge56.7%—
ExtractBench Pro: long-tail document extraction challenges (LlamaIndex)Multimodal12.0%—
Harvey LAB-AA v1.1Agentic6.9%—
STEMBench120: six-domain scientific research progressReasoning2.8%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseEpoch AICC BYVals AINo licence statedLiveBenchNo licence statedARC Prize FoundationTerms restrict reuseEuroEvalMITTerminal-BenchNo licence statedLMArenaCC BY 4.0models.devMIT

History

5 releases
Index79.3 → 78.0
23 Oct 202612 Nov 2026
Output / 1M$10.00 → $10.00
23 Oct 202612 Nov 2026

Questions

How much does GPT-6.1 Sol cost?
$2.00 per 1M input tokens, $10.00 per 1M output tokens.
How good is GPT-6.1 Sol?
Its index is 78.0 out of 100. That is place 5 of 283 ranked models.
What is GPT-6.1 Sol best at?
Math: 78.2 out of 100, place 2 of 168.
How fast is GPT-6.1 Sol?
34 output tokens per second, 4.6 s to the first token.
What is the context window of GPT-6.1 Sol?
1.05M tokens.
Is GPT-6.1 Sol open source?
No. Its weights are not published.
When was GPT-6.1 Sol released?
September 2026.

Badge

GPT-6.1 Sol on Inference360: place 5 of 283, index 78.0
HTML
<a href="https://inference360.ai/models/gpt-6-1-sol"><img src="https://inference360.ai/badges/gpt-6-1-sol.svg" alt="GPT-6.1 Sol on Inference360: place 5 of 283, index 78.0" height="20"></a>
Markdown
[![GPT-6.1 Sol on Inference360: place 5 of 283, index 78.0](https://inference360.ai/badges/gpt-6-1-sol.svg)](https://inference360.ai/models/gpt-6-1-sol)

Compare with

GPT-6 Astra78.0Gemini 4 Argon78.5Claude Opus 5.578.6Muse Wonder Full77.3Claude Sonnet 5.577.2

Lower price

GPT-6.1 Terra77.1 · $5.00/1M

More from OpenAI

All models
GPT-6.1 Sol ProUnrankedGPT-6 Luna ProUnrankedGPT-6 Sol ProUnrankedGPT-6 Sol72.8GPT-6 Luna65.5