Google model product
A thinking model designed for a balance between price and performance. It builds upon Gemini 2.0 Flash with upgraded reasoning, hybrid thinking control, multimodal capabilities (text, image, video, audio input), and a 1M token input context window.
Updated Aug 10, 2026. Default version: Gemini 2.5 Flash
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version Gemini 2.5 Flash.
| FACTS Grounding | 0.9 | 2 | 13 | 91.7% | C | |
| LiveCodeBench v5 | 0.6 | 2 | 9 | 87.5% | C | |
| Global-MMLU-Lite | 0.9 | 3 | 14 | 84.6% | C | |
| Vibe-Eval | 0.7 | 3 | 8 | 71.4% | C | |
| Aider-Polyglot Edit | 0.6 | 5 | 10 | 55.6% | C | |
| MRCR | 0.3 | 7 | 7 | 0.0% | C | |
| Aider-Polyglot | 0.6 | 10 | 22 | 57.1% | C | |
| AIME 2024 | 0.9 | 11 | 53 | 80.8% | C | |
| MMMU | 0.8 | 14 | 63 | 79.0% | C | |
| SimpleQA | 0.3 | 27 | 46 | 42.2% | C | |
| GPQA | 0.8 | 68 | 233 | 71.1% | C | |
| SWE-Bench Verified | 0.6 | 77 | 104 | 26.2% | C | |
| Humanity's Last Exam | 0.1 | 79 | 92 | 14.3% | C | |
| AIME 2025 | 0.7 | 86 | 114 | 24.8% | C |
Preference and agent-evaluation signals from published Arena datasets.
| vision | captioning | 5 | 1277.8 | 582 | N/A | |
| vision style control | captioning | 9 | 1214.6 | 582 | N/A | |
| vision | creative writing | 10 | 1253.1 | 1,745 | N/A | |
| vision style control | creative writing | 13 | 1230.9 | 1,745 | N/A | |
| vision | entity recognition | 18 | 1250.1 | 614 | N/A | |
| text factuality | german | 20 | 1421.0 | 511 | N/A | |
| vision style control | entity recognition | 21 | 1219.9 | 614 | N/A | |
| text factuality | korean | 22 | 1397.3 | 487 | N/A | |
| text factuality | spanish | 33 | 1443.1 | 1,130 | N/A | |
| text factuality | polish | 36 | 1408.0 | 1,133 | N/A | |
| text factuality | french | 44 | 1458.6 | 1,042 | N/A | |
| vision | chinese | 44 | 1285.1 | 2,525 | N/A | |
| vision | creative writing vision | 44 | 1245.9 | 2,711 | N/A | |
| vision | humor | 49 | 1223.6 | 2,132 | N/A | |
| vision style control | creative writing vision | 49 | 1224.6 | 2,711 | N/A | |
| vision style control | humor | 49 | 1204.9 | 2,132 | N/A | |
| vision style control | chinese | 54 | 1256.0 | 2,525 | N/A | |
| vision | english | 56 | 1241.7 | 26,516 | N/A | |
| vision | overall | 57 | 1235.5 | 59,184 | N/A | |
| vision | homework | 58 | 1244.6 | 4,799 | N/A | |
| text factuality | industry legal and government | 59 | 1457.0 | 2,922 | N/A | |
| text | japanese | 60 | 1398.1 | 1,855 | N/A | |
| vision | diagram | 60 | 1246.6 | 8,297 | N/A | |
| vision | ocr | 60 | 1238.1 | 26,513 | N/A | |
| vision style control | homework | 62 | 1237.8 | 4,799 | N/A | |
| vision style control | english | 63 | 1217.4 | 26,516 | N/A | |
| text factuality | creative writing | 65 | 1417.0 | 6,313 | N/A | |
| vision style control | overall | 65 | 1213.9 | 59,184 | N/A | |
| vision style control | diagram | 65 | 1233.3 | 8,297 | N/A | |
| vision style control | ocr | 65 | 1222.0 | 26,513 | N/A | |
| text | korean | 68 | 1388.3 | 2,271 | N/A | |
| text factuality | industry life and physical and social science | 68 | 1462.4 | 6,728 | N/A | |
| text factuality | chinese | 69 | 1468.4 | 2,194 | N/A | |
| text factuality | russian | 70 | 1437.8 | 3,925 | N/A | |
| text factuality | expert | 71 | 1456.7 | 2,952 | N/A | |
| text factuality | industry mathematical | 71 | 1422.1 | 2,116 | N/A | |
| text factuality | industry writing and literature and language | 73 | 1420.5 | 10,351 | N/A | |
| text factuality | math | 73 | 1416.8 | 2,388 | N/A | |
| text factuality | non english | 73 | 1427.4 | 35,497 | N/A | |
| text style control | japanese | 73 | 1393.3 | 1,855 | N/A | |
| text factuality | industry medicine and healthcare | 75 | 1460.5 | 2,594 | N/A | |
| text factuality | industry entertainment and sports and media | 76 | 1408.1 | 8,281 | N/A | |
| text | creative writing | 80 | 1402.2 | 17,205 | N/A | |
| text factuality | overall | 80 | 1437.4 | 68,299 | N/A | |
| text factuality | instruction following | 80 | 1427.5 | 14,424 | N/A | |
| text factuality | longer query | 83 | 1446.4 | 15,624 | N/A | |
| text | german | 85 | 1418.3 | 2,544 | N/A | |
| text | industry writing and literature and language | 85 | 1407.0 | 27,857 | N/A | |
| text | non english | 85 | 1411.4 | 64,703 | N/A | |
| text factuality | exclude ties | 85 | 1431.7 | 48,890 | N/A | |
| text | industry legal and government | 87 | 1432.8 | 8,701 | N/A | |
| text | polish | 87 | 1413.0 | 6,042 | N/A | |
| text | russian | 87 | 1417.3 | 10,080 | N/A | |
| text factuality | hard prompts | 87 | 1455.8 | 38,852 | N/A | |
| text factuality | multi turn | 88 | 1434.5 | 7,762 | N/A | |
| text | industry entertainment and sports and media | 89 | 1394.8 | 22,683 | N/A | |
| text | spanish | 89 | 1418.6 | 2,992 | N/A | |
| text factuality | industry business and management and financial operations | 89 | 1429.9 | 8,441 | N/A | |
| text | longer query | 90 | 1420.1 | 31,653 | N/A | |
| text factuality | hard prompts english | 91 | 1459.4 | 13,711 | N/A | |
| text style control | korean | 91 | 1374.9 | 2,271 | N/A | |
| text | instruction following | 92 | 1405.6 | 33,770 | N/A | |
| text | french | 93 | 1431.8 | 2,363 | N/A | |
| text style control | creative writing | 93 | 1396.5 | 17,205 | N/A | |
| text factuality | english | 94 | 1442.3 | 31,130 | N/A | |
| text factuality | coding | 95 | 1471.2 | 10,765 | N/A | |
| text | expert | 98 | 1425.1 | 7,742 | N/A | |
| text factuality | industry software and it services | 98 | 1461.4 | 20,340 | N/A | |
| text | industry life and physical and social science | 99 | 1433.3 | 20,057 | N/A | |
| text | industry medicine and healthcare | 99 | 1428.9 | 7,669 | N/A | |
| text style control | industry writing and literature and language | 101 | 1403.7 | 27,857 | N/A | |
| text | industry mathematical | 102 | 1421.6 | 6,766 | N/A | |
| text style control | russian | 102 | 1412.3 | 10,080 | N/A | |
| text | exclude ties | 104 | 1404.7 | 88,995 | N/A | |
| text style control | german | 104 | 1407.7 | 2,544 | N/A | |
| text | overall | 106 | 1417.4 | 123,929 | N/A | |
| text | hard prompts | 106 | 1422.5 | 63,505 | N/A | |
| text | chinese | 107 | 1451.9 | 6,386 | N/A | |
| text style control | industry legal and government | 107 | 1427.7 | 8,701 | N/A | |
| text style control | non english | 110 | 1403.3 | 64,703 | N/A | |
| text | industry business and management and financial operations | 112 | 1402.7 | 22,288 | N/A | |
| text style control | industry entertainment and sports and media | 112 | 1387.4 | 22,683 | N/A | |
| text style control | spanish | 113 | 1403.2 | 2,992 | N/A | |
| text style control | polish | 114 | 1399.7 | 6,042 | N/A | |
| text style control | french | 115 | 1425.5 | 2,363 | N/A | |
| text | multi turn | 116 | 1408.3 | 21,583 | N/A | |
| text | english | 117 | 1418.1 | 59,158 | N/A | |
| text | math | 117 | 1410.1 | 7,831 | N/A | |
| text style control | industry mathematical | 118 | 1416.8 | 6,766 | N/A | |
| text | coding | 120 | 1423.4 | 25,761 | N/A | |
| text | hard prompts english | 120 | 1419.5 | 31,975 | N/A | |
| text style control | industry life and physical and social science | 121 | 1430.2 | 20,057 | N/A | |
| text | industry software and it services | 122 | 1425.8 | 43,433 | N/A | |
| text style control | instruction following | 122 | 1400.9 | 33,770 | N/A | |
| text style control | chinese | 123 | 1442.4 | 6,386 | N/A | |
| text style control | longer query | 123 | 1418.4 | 31,653 | N/A | |
| text style control | expert | 125 | 1422.9 | 7,742 | N/A | |
| text style control | math | 125 | 1405.4 | 7,831 | N/A | |
| text style control | overall | 126 | 1410.2 | 123,929 | N/A | |
| text style control | exclude ties | 126 | 1395.5 | 88,995 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| QiHang | gemini-2.5-flash | global | $0.09 | $0.71 | 1M | |
| Jiekou.AI | gemini-2.5-flash | global | $0.27 | $2.25 | 1M | |
| Cortecs | gemini-2.5-flash | global | $0.299 | $2.49 | 1M | |
| gemini-2.5-flash | global | $0.30 | $2.5 | 1M | ||
| NanoGPT | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| LLM Gateway | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| 302.AI | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| Helicone | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| FrogBot | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| SAP AI Core | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| Abacus | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| Auriko | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| AIHubMix | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| Vertex | gemini-2.5-flash | global | $0.30 | $2.5 | 1M | |
| Modelis | gemini-2.5-flash | global | $0.30 | $2.5 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| Gemini 2.5 Flash | 63.5 | N/A | 1M | 65.5K | No | Proprietary |
A concise description based on the published model registry.
A thinking model designed for a balance between price and performance. It builds upon Gemini 2.0 Flash with upgraded reasoning, hybrid thinking control, multimodal capabilities (text, image, video, audio input), and a 1M token input context window.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Gemini 2.5 Flash.
Gemini 2.5 Flash's default version was released on May 20, 2025.
Gemini 2.5 Flash's official API price is $0.30 per million input tokens and $2.5 per million output tokens via Google. The lowest tracked third-party offer starts at $0.09 input and $0.71 output via QiHang.
Gemini 2.5 Flash is published under Google in the model registry.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
16 published provider offerings are linked to the default version.
Nearby ranked alternatives include Qwen3 235B A22B Thinking, GLM 5.1, Claude Sonnet 3.7.