Gemini 3.5 Flash
Gemini 3.5 Flash is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.
Use it for
- Teams evaluating coding, rag, and agents
- Workloads that can use a 1.05m context window
- Buyers comparing 4 tracked provider routes
Do not use it for
- Workloads where another current model has stronger sourced task evidence
- Family
- Gemini 3.5
- Released
- 2026-05-19
- Context
- 1.05m
- Max output
- 65,536
- Architecture
- Decoder Only
- Knowledge cutoff
- 2025-01
- Specialization
- general
- Openness
- Proprietary
- License
- ProprietaryCommercial use: conditional
- Weights
- Not released
- Code
- Unknown
- Training
- Pretrained
Cheapest of 4 routes · GCP Vertex AI · cache read $0.150
About
Gemini 3.5 Flash is Google DeepMind's generally available Flash model for sustained frontier-level performance on agentic and coding tasks. It supports multimodal inputs, native thinking, tool and function calling, structured outputs, code execution, search grounding, batch processing, and long contexts up to 1M tokens.
Top use-case fit: coding, agents, and build tasks
Coding
Q/$ D3 relevant benchmarks in the decision map.
RAG
Included by capability and metadata signals in the decision map.
Agents
Q/$ C1 relevant benchmark in the decision map.
Provider price ladder
Compare all 4Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Batch in / out | Cache | Route |
|---|---|---|---|---|---|
| GCP Vertex AI | $1.50 | $9.00 | $0.750 / $4.50 | read $0.150 | Serverless |
| Google AI Studio | $1.50 | $9.00 | $0.750 / $4.50 | read $0.150 | Serverless |
| OpenRouter | $1.50 | $9.00 | - | - | Serverless |
| Vercel AI Gateway | $1.50 | $9.00 | - | read $0.150 | Serverless |
Available via routers & gateways(13)
LiteLLM
GatewayOpen-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.
OpenRouter
HybridUnified hybrid gateway to 400+ models from 60+ providers via a single OpenAI-compatible API, with optional auto-routing that selects the best model per prompt.
Portkey
GatewayProduction AI gateway routing to 1,600+ LLMs with failover, load balancing, semantic caching, and guardrails; Apache 2.0 core is fully self-hostable with the complete feature set.
AIRouter
RouterCommercial LLM router that analyzes incoming requests and routes to the optimal model for cost/quality/latency via a drop-in OpenAI-compatible API, with a privacy-preserving embedding mode that avoids sending prompt content.
Helicone
GatewayObservability-first AI gateway with routing, caching, rate limiting, and request tracing; Apache 2.0 open-source core with a managed hosted tier for logging and analytics.
Kong AI Gateway
GatewayMulti-LLM AI gateway built on Kong Gateway 3.x, adding semantic routing, load balancing, guardrails, and MCP traffic analytics as plugins over Kong's existing API management platform.
Capabilities
Benchmark peer barsfor Coding
Benchmark scores(14)
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| MMMU Pro | 88.3 | Vals.ai standardized CoT harness, 4-option, Pass@1, temp=0Observed 2026-06-02 | — | Source |
| SWE-bench Pro | 55.1 | SWE-bench Pro Public (pass@1)Observed 2026-06-07 | — | Source |
| Terminal-Bench | 76.2 | —Observed 2026-05-19 | — | Source |
| Massive Multi-discipline Multimodal Understanding | 83.6 | —Observed 2026-05-19 | — | Source |
| HumanEval | 92.0 | —Observed 2026-05-25 | — | Source |
| ARC-AGI-2 | 72.1 | ARC-AGI-2 (accuracy%)Observed 2026-06-07 | — | Source |
| Google-Proof Q&A | 92.2 | GPQA Diamond (accuracy)Observed 2026-06-07 | — | Source |
| Humanity's Last Exam | 40.2 | HLE (accuracy)Observed 2026-06-07 | — | Source |
| MCP-Atlas | 83.6 | MCP-Atlas (accuracy%)Observed 2026-06-07 | — | Source |
| SWE-bench Verified | 78.0 | SWE-bench Verified (pass@1)Observed 2026-06-07 | — | Source |
| Terminal-Bench 2.0 | 76.2 | Terminal-Bench 2.0 (accuracy%)Observed 2026-06-07 | — | Source |
| CursorBench | 49.8 | CursorBench 3.1Observed 2026-06-30 | Configuration: Gemini 3.5 Flash (single reported configuration) Harness: CursorBench 3.1 Evaluator: Cursor Confidence: confirmed Notes: Cursor published one CursorBench 3.1 configuration for this model; no cross-effort selection was needed. | Source |
| GeneBench-Pro | 8.1 | highObserved 2026-06-30 | — | Source |
| CursorBench | 48.8 | CursorBench 3.2Observed 2026-07-18 | Configuration: Gemini 3.5 Flash Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $2.20 Tokens/task: 46,702 Steps/task: 77 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful. | Source |
Migration checks
No linked migration route is available for this model yet.
API versions
gemini-3.5-flashRankings & picks(5)
Compare Gemini 3.5 Flash with other models
Comparison and alternatives
Browse all comparisons →Cheapest of 4 routes · GCP Vertex AI · cache read $0.150