# Google Gemini 3.5 Flash Lite As of October 21, 2026, Google Gemini 3.5 Flash Lite ranks #74 of 123 models overall with 49.8 ±9.8 on the LLMBenchmark Index. Gemini 3.5 Flash Lite ranks #55 of 96 in coding, its strongest category. Google Gemini 3.5 Flash Lite costs $0.30 per million input tokens and $2.50 per million output tokens. Gemini 3.5 Flash Lite was released on July 21, 2026. Page: https://llmbenchmark.ai/models/gemini-3-5-flash-lite Data from October 21, 2026. Every figure below names the board it was read from and the day. - Maker: Google - Released: July 21, 2026 - Price per million input tokens: $0.30 - Price per million output tokens: $2.50 - Context window: 1.05M tokens - Overall rank: 74 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 49.8 ±9.8 (90% margin, from 5 of 7 inputs) - Artificial Analysis Intelligence Index: 22.2 - Vibe Coding Index: 49.0 (from 3 of 6 coding benchmarks) Benchmark results (85 of 195): - LMArena WebDev: 1435 Elo (source: LMArena WebDev, read October 21, 2026) - GPQA Diamond: 83.8% (source: GPQA (Vals AI), read October 21, 2026) - MMLU Pro: 85.8% (source: MMLU Pro (Vals AI), read October 21, 2026) - AA Intelligence Index: 22.2 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - AA Coding Index: 49.3 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - AA Agentic Index: 14.3 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - LiveCodeBench (Vals AI): 79.0% (source: Vals AI boards, read October 21, 2026) - SWE-bench Verified (Vals AI): 75.0% (source: Vals AI boards, read October 21, 2026) - Vibe Code Bench (Vals AI): 37.2% (source: Vals AI boards, read October 21, 2026) - Code migration (Vals AI): 6.2% (source: Vals AI boards, read October 21, 2026) - ProgramBench (Vals AI): 0.0% (source: Vals AI boards, read October 21, 2026) - Terminal-Bench 2.1 (Vals AI): 50.2% (source: Vals AI boards, read October 21, 2026) - CyberBench (Vals AI): 59.1% (source: Vals AI boards, read October 21, 2026) - Finance Agent (Vals AI): 47.4% (source: Vals AI boards, read October 21, 2026) - Tax Agent Bench (Vals AI): 7.2% (source: Vals AI boards, read October 21, 2026) - Legal Research Bench (Vals AI): 13.9% (source: Vals AI boards, read October 21, 2026) - Excel Modeling Benchmark (Vals AI): 43.2% (source: Vals AI boards, read October 21, 2026) - Public Benefits Bench (Vals AI): 51.8% (source: Vals AI boards, read October 21, 2026) - MMMU Pro (Vals AI): 83.6% (source: Vals AI boards, read October 21, 2026) - SAGE (Vals AI): 47.3% (source: Vals AI boards, read October 21, 2026) - LegalBench (Vals AI): 84.1% (source: Vals AI boards, read October 21, 2026) - CorpFin (Vals AI): 60.7% (source: Vals AI boards, read October 21, 2026) - TaxEval (Vals AI): 72.6% (source: Vals AI boards, read October 21, 2026) - MortgageTax (Vals AI): 68.7% (source: Vals AI boards, read October 21, 2026) - MedCode (Vals AI): 43.5% (source: Vals AI boards, read October 21, 2026) - MedScribe (Vals AI): 70.9% (source: Vals AI boards, read October 21, 2026) - GPQA Diamond (Artificial Analysis run): 83.8% (source: Artificial Analysis, read October 21, 2026) - Humanity's Last Exam (Artificial Analysis run): 18.8% (source: Artificial Analysis, read October 21, 2026) - SciCode: 41.3% (source: Artificial Analysis, read October 21, 2026) - Long context reasoning (AA-LCR): 76.0% (source: Artificial Analysis, read October 21, 2026) - WeirdML: 39.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - APEX-Agents: 29.3% (source: Epoch AI benchmarking hub, read October 21, 2026) - GPQA Diamond (Epoch AI run): 83.3% (source: Epoch AI benchmarking hub, read October 21, 2026) - ARC-AGI: 53.5% (source: Epoch AI benchmarking hub, read October 21, 2026) - ARC-AGI-2: 10.3% (source: Epoch AI benchmarking hub, read October 21, 2026) - Chess puzzles: 21.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - Mystery game puzzles: 19.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - OTIS Mock AIME: 71.1% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierMath, tiers 1 to 3: 26.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierMath, tier 4: 0.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - ProofBench: 13.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - DTBench: 83.5% (source: Epoch AI benchmarking hub, read October 21, 2026) - LMCA: 37.6% (source: Epoch AI benchmarking hub, read October 21, 2026) - ALE-Bench: 765.3 (source: ALE-Bench (Sakana AI), read October 21, 2026) - Epoch Capabilities Index: 144.9 (source: Epoch AI benchmarking hub, read October 21, 2026) - LMArena Text: 1435 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: coding: 1453 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: hard prompts: 1441 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: math: 1434 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: instruction following: 1419 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: creative writing: 1421 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: 1269 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: English: 1438 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: non-English: 1427 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Chinese: 1468 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: German: 1454 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Japanese: 1426 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Korean: 1407 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: French: 1451 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Spanish: 1441 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Russian: 1443 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Polish: 1437 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: multi-turn: 1445 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: longer queries: 1435 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: expert: 1434 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: hard prompts in English: 1440 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: software and IT: 1451 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: mathematical work: 1433 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: science: 1451 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: medicine and healthcare: 1438 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: legal and government: 1431 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: business and finance: 1423 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: writing and language: 1428 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: entertainment and media: 1414 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena WebDev: React: 1408 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena WebDev: from an image: 1448 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: English: 1269 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: Chinese: 1311 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: reading text: 1280 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: diagrams: 1284 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: homework: 1291 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: humor: 1287 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Vision: creative writing: 1260 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - SciCode: 6.2% (source: Kaggle Benchmarks, read October 21, 2026) - FACTS Benchmark Suite: 53.5% (source: Kaggle Benchmarks, read October 21, 2026)