BlogLLM Benchmark Scores as Published, Side by Side
Gemini 3.5 Flash: Benchmark Scores as Published
Every result the catalog holds for Gemini 3.5 Flash: 21 tests in 11 areas, from 5 pages. 18 printed by Google itself, the rest by the company named beside each number. Each number is as printed; nothing is ranked or rescored.
Developer GoogleReleased 2026-05-19Context window 1MMax output 64KInput Text, image, audio, videoOutput TextWeights ClosedData version 2026103001, as of 2026-10-30
Knowledge Work
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| AutomationBench | 14.5% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15 | - |
| GDPval-AA v2 | 1349 | Google, Gemini 3.6 Flash model cardchecked 2026-10-15; printed on 1 more page | - |
| GDPval-AA (Elo) | 1656 | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | - |
| Vals Finance Agent v2 | 57.9% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | - |
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| SWE-Bench Pro | 55.1% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | - |
| Terminal-Bench 2.1 | 76.2% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | Harness: Terminus-2 |
| DeepSWE v1.1 | 37% | Google, Gemini 3.6 Flash model cardchecked 2026-10-15 | - |
ML Engineering
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MLE-bench | 49.7% | Google, Gemini 3.6 Flash model cardchecked 2026-10-15 | - |
Science and Math
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| GeneBench Pro | 8.14% | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15 | - |
Long Context
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MRCR v2 8-needle 128K | 77.3% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | Context: 128k (average) |
| MRCR v2 8-needle 1M (pointwise) | 26.6% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | Context: 1M (pointwise) |
Computer Use
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| OSWorld-Verified | 78.4% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | - |
Multimodal Understanding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| CharXiv Reasoning (no tools) | 84.2% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15; printed on 1 more page | Tools: none |
| MMMU-Pro | 83.6% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | Tools: none |
| CharXiv Reasoning (with tools) | 84.9% | Google, Gemini 3.6 Flash model cardchecked 2026-10-15 | Tools: enabled; exact tools in source |
General Knowledge and Reasoning
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Humanity's Last Exam (no tools) | 40.2% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | - |
| ARC-AGI-2 | 72.1% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | - |
Tool Use and Workflows
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Toolathlon | 56.5% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | - |
| MCP Atlas | 83.6% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | - |
Engineering and Design
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Blueprint-Bench 2 (normalized score) | 33.6% | Google, Gemini 3.5 Flash model cardchecked 2026-10-15 | - |
Preference and Composite Rankings
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Artificial Analysis Intelligence Index v4.1 | 50.2 | OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15 | - |
Price
API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked. Cost per task is measured by Vals AI.
| Price | Value | Printed by | Setup as printed |
|---|---|---|---|
| Cost per taskVals AI, USD, lower is better | $2.54 | Vals AI, Vibe Code Bench v1.1 leaderboardchecked 2026-10-27 | Harness: OpenHands |
More from Google
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.