What other models does Google offer?

Google offers a broad range of Gemini models that span from budget-optimized flash variants to flagship reasoning models. Here are some other models Google supports:

• Gemini 3 Pro: Gemini 3 Pro is Google's premium reasoning model in the Gemini 3 generation, priced at $2.00 per million input tokens and $12.00 per million output tokens, ranked #30 out of 150 models on the Artificial Analysis Intelligence Index, and suited for the most demanding analytical and agentic tasks where cost is secondary to accuracy

• Gemini 2.5 Pro: Gemini 2.5 Pro is Google's stable flagship from the prior generation, priced at $1.25 per million input tokens and $10.00 per million output tokens, supporting extended reasoning over a 1M token context window and a reliable production alternative to preview-stage Gemini 3 models

• Gemini 2.5 Flash: Gemini 2.5 Flash is a stable mid-tier model producing 203.2 tokens per second at $0.30 per million input tokens and $2.50 per million output tokens, positioned as a proven general-purpose option with above-average intelligence for its price range

• Gemini 2.5 Flash Lite: Gemini 2.5 Flash Lite is the fastest and least expensive model in Google's current lineup, priced at $0.10 per million input tokens and $0.40 per million output tokens, delivering the highest output speed across all evaluated models at 259.8 tokens per second and designed for high-throughput tasks where cost and latency dominate requirements

How does Gemini 3 Flash differ from Google's other models?

Gemini 3 Flash occupies the cost-efficient tier of the Gemini 3 series, targeting teams that want the newer generation's capabilities without paying flagship prices.

• Pricing: Input costs $0.50 per million tokens and output costs $3.00 per million tokens. That is one-quarter the input cost of Gemini 3 Pro at $2.00 per million, and less than half the output cost of Gemini 2.5 Pro at $10.00 per million

• Speed: Output speed is 158.4 tokens per second, which places it below Gemini 2.5 Flash at 203.2 tokens per second and well below Gemini 2.5 Flash Lite at 259.8 tokens per second. Time to first token is 22.03 seconds, reflecting latency characteristics typical of more capable models

• Context window: Supports 1M tokens of input context, matching Gemini 3 Pro and Gemini 2.5 Pro

• Intelligence Index: Ranked #20 out of 71 non-reasoning models on the Artificial Analysis Intelligence Index, placing it above average in intelligence for its tier among non-reasoning models

• Reasoning mode: Gemini 3 Flash in its standard (non-reasoning) configuration does not generate extended chain-of-thought, unlike Gemini 3 Pro. This keeps latency lower at the cost of reduced accuracy on multi-step reasoning tasks

Gemini 3 Flash is the right default for teams that want the Gemini 3 generation's intelligence gains at a price point closer to Gemini 2.5 Flash, particularly for tasks that don't require extended reasoning.

What models should I consider using alongside Gemini 3 Flash?

No single model is optimal for every task. Here are models worth pairing with Gemini 3 Flash depending on what your product needs:

• Gemini 3 Pro: For the subset of requests within your product that require deep reasoning, scientific analysis, or complex multi-step agentic behavior, route those specifically to Gemini 3 Pro while keeping Gemini 3 Flash as the default for everything else

• Gemini 2.5 Flash Lite: For bulk preprocessing tasks like input classification, entity extraction, or prompt filtering before a Gemini 3 Flash call, Gemini 2.5 Flash Lite at $0.10 per million input tokens reduces overall pipeline cost without affecting the quality of the Gemini 3 Flash output that follows

• Claude Haiku 3.5 (Anthropic): For structured output tasks where strict JSON schema adherence and predictable token counts matter, Claude Haiku 3.5 provides a cost-comparable cross-provider alternative with strong instruction-following for formatting-intensive workloads

• GPT-4o mini (OpenAI): For workloads already built on the OpenAI SDK where switching provider logic is minimal, GPT-4o mini offers a comparable price-tier option and serves as a reliable fallback if Google's API experiences an outage

• Mistral Small (Mistral AI): For European data residency requirements or workloads where a European-hosted provider is preferable, Mistral Small covers many of the same general-purpose instruction-following use cases as Gemini 3 Flash at a similar pricing tier

What are the challenges of using Gemini 3 Flash in my product?

Like any production LLM, Gemini 3 Flash comes with tradeoffs worth planning for:

• High time to first token: At 22.03 seconds time to first token, Gemini 3 Flash is on the slower end of response initiation for its price tier. Interactive applications like chat interfaces or streaming completions will feel sluggish to users without careful handling, such as showing a typing indicator or streaming tokens as they arrive

• Preview status and deprecation risk: Gemini 3 Flash is a preview-stage model. Rate limits, availability guarantees, and API behavior may differ from stable models, and preview models can be deprecated on shorter timelines than production releases

• Provider dependency: Concentrating traffic on Google's API means that quota restrictions or regional outages affect every workflow relying on Gemini 3 Flash. Diversifying to at least one fallback provider reduces that exposure

• Cost at scale: At $3.00 per million output tokens, output costs compound quickly at high volumes. Applications with long completions or multi-turn conversation histories should track output token usage closely against projected spend

• No native extended reasoning in standard mode: Gemini 3 Flash (non-reasoning) does not generate chain-of-thought by default. Tasks that benefit from multi-step problem decomposition, such as competitive math or complex code debugging, may underperform compared to Gemini 3 Pro or other reasoning-enabled models

Why should I use Merge Gateway to route LLM requests with Gemini 3 Flash and every other model?

Using Gemini 3 Flash through Merge Gateway gives you access to the model itself and the infrastructure layer around it:

• Intelligent routing and automatic failover: Merge routes around Google outages automatically. Routing policies based on cost, latency, or quality can reduce spend by 40-60% without touching your application code, which is especially useful given Gemini 3 Flash's preview-stage availability constraints

• One API, every provider: Access Gemini 3 Flash and every other major LLM through a single endpoint and API key. Change providers by swapping the model string, with no application code changes required

• Cost governance: Set hard or soft project budgets so Gemini 3 Flash spend stays within plan. Every request is attributed to a model, project, and tag in a unified billing dashboard across all providers

• Build Your Own Router: Define what "best" means for your traffic by selecting from curated ML benchmarks or adding your own eval scores. The router scores each available model against your weights and picks the winner per request, with a plain-language explanation of every decision

• Security and compliance controls: Apply DLP rules and prompt injection protection before every request reaches Google. Enforce per-project model and region policies without adding that logic to your application

How can I start using Merge Gateway to route requests with Gemini 3 Flash?

Getting Gemini 3 Flash running through Merge Gateway takes a few minutes:‍

1. Create an account and get your API key from the dashboard.

2. Install the Merge Gateway SDK: run pip install merge-gateway-sdk (Python) or npm install merge-gateway-sdk (Node). Alternatively, if you're already using the OpenAI SDK, set base_url = "https://api-gateway.merge.dev/v1/openai" and your existing code works as-is.

3. Make your first request using the provider/model format. For Gemini 3 Flash, the model string is google/gemini-3-flash. Swap the model string to route to any other provider without changing anything else.

4. Configure a routing policy in the dashboard to set failover behavior, cost limits, and optimization strategy. Your first policy can be as simple as naming Gemini 3 Flash as primary with one fallback.

Full setup instructions and SDK references are in the Merge Gateway docs.