Claude Opus 5 vs GPT-6.1 Sol

Anthropic against OpenAI. Prices, scores and speed below are live.

  • GPT-6.1 Sol costs 60% less on blended price ($6.00 against $15.00 per 1M tokens, 3 input to 1 output).
  • GPT-6.1 Sol scores higher on the Artificial Analysis Intelligence Index: 54.6 against 44.2.
  • On the 38 benchmarks both have scores on, Claude Opus 5 leads on 17 and GPT-6.1 Sol on 20, with 1 tied.
  • GPT-6.1 Sol reads more context: 1.1M tokens against 1M.
  • GPT-6.1 Sol is newer: listed Sep 29, 2026, against Jul 24, 2026.

Side by side

Claude Opus 5GPT-6.1 Sol
LabAnthropicOpenAI
ListedJul 24, 2026Sep 29, 2026
Input per 1M tokens$7.50$3.00
Output per 1M tokens$37.50$15.00
Cached input per 1M$0.75$0.15
Blended (3:1) per 1M$15.00$6.00
Context window1M1.1M
Max output128K128K
Knowledge cutoff--
Intelligence index44.254.6
Coding index78-
ReasoningYesYes
Reads imagesYesYes
Tool useYesYes
Structured outputYesYes

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

BenchmarkClaude Opus 5GPT-6.1 Sol
AIME93.3%Epoch AI100%Epoch AI
APEX-Agents65.8%Epoch AI60%Epoch AI
ARC-AGI-197.5%Epoch AI98.5%Epoch AI
ARC-AGI-290.4%Epoch AI94.2%Epoch AI
Artificial Analysis Intelligence Index44.2Artificial Analysis and OpenRouter54.6Artificial Analysis and OpenRouter
Chess Puzzles43%Epoch AI68%Epoch AI
Design Arena: 3D1324 EloDesign Arena1425 EloDesign Arena
Design Arena: Android Native (agentic)1246 EloDesign Arena1202 EloDesign Arena
Design Arena: Code1318 EloDesign Arena1354 EloDesign Arena
Design Arena: Data Viz1327 EloDesign Arena1281 EloDesign Arena
Design Arena: Game Dev1338 EloDesign Arena1407 EloDesign Arena
Design Arena: Mobile Apps (agentic)1348 EloDesign Arena1228 EloDesign Arena
Design Arena: UI Component1340 EloDesign Arena1342 EloDesign Arena
Design Arena: Web Apps (agentic)1253 EloDesign Arena1264 EloDesign Arena
Design Arena: Website1304 EloDesign Arena1308 EloDesign Arena
DTBench97.9%Epoch AI96.5%Epoch AI
Earthborne Rangers (EBR-bench)45.7Epoch AI54.3Epoch AI
FrontierCode53.4Epoch AI50.2Epoch AI
FrontierMath Tier 4 (v2)73.2%Epoch AI100%Epoch AI
FrontierMath Tiers 1-3 (v2)85.6%Epoch AI93.7%Epoch AI
GPQA Diamond87.9%Artificial Analysis and OpenRouter94.4%Artificial Analysis and OpenRouter
LMArena (Coding)1532 EloLMArena1486 EloLMArena
LMArena (Overall)1506 EloLMArena1446 EloLMArena
LMArena: Creative Writing1494 EloLMArena1420 EloLMArena
LMArena: Hard Prompts1525 EloLMArena1467 EloLMArena
LMArena: Longer Query1516 EloLMArena1460 EloLMArena
LMArena: Math1543 EloLMArena1495 EloLMArena
LMArena: Multi-Turn1499 EloLMArena1452 EloLMArena
LMCA64.5Epoch AI62.8Epoch AI
MirrorCode (ML, +Private, 2L)87.8Epoch AI61.1Epoch AI
Mystery Game Puzzles37%Epoch AI80%Epoch AI
PostTrainBench35.7Epoch AI38.8Epoch AI
ProofBench99%Epoch AI99%Epoch AI
SciCode56.4%Epoch AI55.8%Epoch AI
SimpleBench80.6%Epoch AI82.9%Epoch AI
SimpleQA Verified59.9%Epoch AI73.9%Epoch AI
tau-bench (Airline)78.7%Artificial Analysis and OpenRouter70%Artificial Analysis and OpenRouter
WebDev Arena1687 EloEpoch AI1759 EloEpoch AI

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See Anthropic models, OpenAI models, every LLM's price and our sources.