# OpenAI GPT-5 Mini As of October 21, 2026, OpenAI GPT-5 Mini ranks #86 of 123 models overall with 43.7 ±9.0 on the LLMBenchmark Index. GPT-5 Mini ranks #68 of 108 in math, its strongest category. OpenAI GPT-5 Mini costs $0.25 per million input tokens and $2.00 per million output tokens. GPT-5 Mini was released on August 7, 2025. Page: https://llmbenchmark.ai/models/gpt-5-mini Data from October 21, 2026. Every figure below names the board it was read from and the day. - Maker: OpenAI - Released: August 7, 2025 - Price per million input tokens: $0.25 - Price per million output tokens: $2.00 - Context window: 400K tokens - Overall rank: 86 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 43.7 ±9.0 (90% margin, from 6 of 7 inputs) - Artificial Analysis Intelligence Index: 20.6 - Vibe Coding Index: 34.2 (from 2 of 6 coding benchmarks) Benchmark results (101 of 195): - SWE-bench Verified: 59.8% (source: SWE-bench (swebench.com), read October 21, 2026) - SWE-bench Multilingual: 39.7% (source: SWE-bench (swebench.com), read October 21, 2026) - GPQA Diamond: 80.3% (source: GPQA (Vals AI), read October 21, 2026) - AIME: 91.5% (source: AIME (Vals AI), read October 21, 2026) - MMLU Pro: 82.2% (source: MMLU Pro (Vals AI), read October 21, 2026) - AA Intelligence Index: 20.6 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - Design Arena: websites: 1125 Elo (source: Design Arena, read October 21, 2026) - Design Arena: UI components: 1106 Elo (source: Design Arena, read October 21, 2026) - Design Arena: game dev: 1130 Elo (source: Design Arena, read October 21, 2026) - Design Arena: data viz: 1130 Elo (source: Design Arena, read October 21, 2026) - Design Arena: 3D: 1037 Elo (source: Design Arena, read October 21, 2026) - Design Arena: all code categories: 1115 Elo (source: Design Arena, read October 21, 2026) - Design Arena: SVG: 1091 Elo (source: Design Arena, read October 21, 2026) - Design Arena: ASCII art: 1125 Elo (source: Design Arena, read October 21, 2026) - LiveCodeBench (Vals AI): 86.6% (source: Vals AI boards, read October 21, 2026) - SWE-bench Verified (Vals AI): 60.8% (source: Vals AI boards, read October 21, 2026) - Vibe Code Bench (Vals AI): 14.2% (source: Vals AI boards, read October 21, 2026) - MMMU Pro (Vals AI): 78.9% (source: Vals AI boards, read October 21, 2026) - MATH 500 (Vals AI): 94.8% (source: Vals AI boards, read October 21, 2026) - MGSM (Vals AI): 92.6% (source: Vals AI boards, read October 21, 2026) - SAGE (Vals AI): 43.0% (source: Vals AI boards, read October 21, 2026) - LegalBench (Vals AI): 81.8% (source: Vals AI boards, read October 21, 2026) - CaseLaw (Vals AI): 68.5% (source: Vals AI boards, read October 21, 2026) - CorpFin (Vals AI): 60.2% (source: Vals AI boards, read October 21, 2026) - TaxEval (Vals AI): 75.2% (source: Vals AI boards, read October 21, 2026) - MortgageTax (Vals AI): 66.9% (source: Vals AI boards, read October 21, 2026) - MedQA (Vals AI): 96.1% (source: Vals AI boards, read October 21, 2026) - MedCode (Vals AI): 43.0% (source: Vals AI boards, read October 21, 2026) - MedScribe (Vals AI): 80.6% (source: Vals AI boards, read October 21, 2026) - AA Math Index: 90.7 (source: Artificial Analysis, read October 21, 2026) - GPQA Diamond (Artificial Analysis run): 82.8% (source: Artificial Analysis, read October 21, 2026) - Humanity's Last Exam (Artificial Analysis run): 21.5% (source: Artificial Analysis, read October 21, 2026) - MMLU Pro (Artificial Analysis run): 83.7% (source: Artificial Analysis, read October 21, 2026) - LiveCodeBench (Artificial Analysis run): 83.8% (source: Artificial Analysis, read October 21, 2026) - SciCode: 39.0% (source: Artificial Analysis, read October 21, 2026) - Terminal-Bench Hard (Artificial Analysis run): 33.3% (source: Artificial Analysis, read October 21, 2026) - IFBench: 75.4% (source: Artificial Analysis, read October 21, 2026) - Long context reasoning (AA-LCR): 72.3% (source: Artificial Analysis, read October 21, 2026) - τ²-Bench Telecom: 71.1% (source: Artificial Analysis, read October 21, 2026) - AIME 2025 (Artificial Analysis run): 90.7% (source: Artificial Analysis, read October 21, 2026) - Terminal-Bench: 34.8% (source: Epoch AI benchmarking hub, read October 21, 2026) - WeirdML: 52.7% (source: Epoch AI benchmarking hub, read October 21, 2026) - GPQA Diamond (Epoch AI run): 75.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - ARC-AGI: 54.3% (source: Epoch AI benchmarking hub, read October 21, 2026) - ARC-AGI-2: 4.4% (source: Epoch AI benchmarking hub, read October 21, 2026) - Chess puzzles: 24.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - Mystery game puzzles: 10.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - OTIS Mock AIME: 86.7% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierMath, tiers 1 to 3: 46.7% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierMath, tier 4: 12.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - ProofBench: 9.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - MATH level 5: 97.8% (source: Epoch AI benchmarking hub, read October 21, 2026) - SimpleQA Verified: 21.6% (source: Epoch AI benchmarking hub, read October 21, 2026) - DTBench: 80.5% (source: Epoch AI benchmarking hub, read October 21, 2026) - LMCA: 34.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - Fiction.LiveBench: 69.4% (source: Epoch AI benchmarking hub, read October 21, 2026) - Creative writing (Lech Mazur): 83.1% (source: Epoch AI benchmarking hub, read October 21, 2026) - VPCT: 40.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - Vending-Bench 2: -$31 (source: Epoch AI benchmarking hub, read October 21, 2026) - Epoch Capabilities Index: 145.3 (source: Epoch AI benchmarking hub, read October 21, 2026) - LMArena Text: 1373 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: coding: 1406 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: hard prompts: 1380 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: math: 1378 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: instruction following: 1357 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: creative writing: 1325 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: 1202 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: English: 1379 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: non-English: 1363 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Chinese: 1386 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: German: 1366 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Japanese: 1341 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Korean: 1308 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: French: 1386 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Spanish: 1354 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Russian: 1362 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Polish: 1378 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: multi-turn: 1363 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: longer queries: 1355 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: expert: 1379 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: hard prompts in English: 1383 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: software and IT: 1394 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: mathematical work: 1380 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: science: 1371 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: medicine and healthcare: 1380 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: legal and government: 1375 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: business and finance: 1358 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: writing and language: 1344 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: entertainment and media: 1340 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: English: 1216 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: Chinese: 1214 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: reading text: 1199 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: diagrams: 1202 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: homework: 1210 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: humor: 1205 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: captioning: 1217 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: recognizing things: 1226 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: creative writing: 1181 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - Berkeley Function Calling Leaderboard (BFCL): 55.5% (source: Berkeley Function Calling Leaderboard, read October 21, 2026) - Kagi LLM Benchmark: 70.3% (source: Kagi LLM Benchmarking Project, read October 21, 2026) - FACTS Benchmark Suite: 38.6% (source: Kaggle Benchmarks, read October 21, 2026)