Gemini 3.5 Flash
Frontier agentic coding at sub-flagship economics.
Context
1M
Max output
64K
Input /1M
$1.50
Output /1M
$9.00
Live pricing via OpenRouter
Best for
- High-volume agentic workloads where cost compounds
- MCP-heavy tool-use pipelines (83.6% MCP Atlas)
- Long-horizon coding agents that need 1M context cheaply
Watch out
Pure abstract reasoning still trails Claude Opus 4.6 (ARC-AGI-2). Vendor benchmarks pending independent reproduction.
For creators. Best price/quality for content-generation agents and bulk drafting loops.
Benchmarks
| terminal bench 2 1 | 76.2 |
| mcp atlas | 83.6 |
| gdpval aa | 1656 |
Capabilities
- Frontier agentic coding at sub-flagship economics
- 1M token context window
- MCP-native tool use (83.6% MCP Atlas)
- Full multimodal: text, vision, audio, video
Compare Gemini 3.5 Flash
Gemini 3.5 Flash vs Claude Opus 4.6
Different tiers, different jobs. Flash wins cost-sensitive agentic coding (76.2% Terminal-Bench 2.1); Opus 4.6 wins high-stakes reasoning. Note: Opus 4.6 was superseded by Opus 4.8 in May 2026; see Opus 4.8 vs GPT-5.5 for that May 2026 matchup.
Gemini 3.5 Flash vs GPT-5.2 Pro
Flash for cost-efficient agentic coding at scale; GPT-5.2 Pro for voice-native and broad multimodal apps. Note: GPT-5.2 Pro is superseded by GPT-5.5 — see Grok 4.3 vs GPT-5.5 or Opus 4.8 vs GPT-5.5 for the April–May 2026 matchups.
Claude Sonnet 4.5 vs Gemini 3.5 Flash
Gemini 3.5 Flash edges ahead on agentic-coding benchmarks and cost; Claude Sonnet 4.5 remains a proven production workhorse. Note: newer Anthropic mid-tier models have shipped since (Sonnet 4.6 in February 2026, Sonnet 5 in June 2026, Sonnet 5.5 on 28 September 2026).