Gemini 3.1 Flash Lite

Google Vertex AI chat Deprecates in 184d (May 7, 2027)

Gemini 3.1 Flash Lite is a Google Vertex AI chat model.It supports a 1,048,576-token context windowwith up to 65,536 output tokens.Input is priced at $0.250/M tokens and output at $1.50/M tokens. Capabilities include function calling, vision, reasoning, audio input. Route Gemini 3.1 Flash Lite via Future AGI's Agent Command Center for unified observability, caching, and 15 routing strategies including cost-optimized fallback.

Pricing source: litellm Last verified: Oct 22, 2026 View source ↗
Cost calculator

Estimate Gemini 3.1 Flash Lite spend

Pick a workload, fine-tune the sliders, and see the monthly bill.

~3K in / ~400 out · 5K req/day
3,000
01,048,576
400
065,536
5,000
01,000,000
cached @ $0.0250/M
Per request
$0.001350
in $0.000750 · out $0.000600
Per day
$6.75
5,000 requests
Per month
$205
152,188 requests

Estimate uses $0.2500/M input · $1.50/M output. Provider pricing changes. Production costs vary with retries, streaming overhead, and tool-call rounds.
Want this for free? Cache + route via Agent Command Center — first 100K requests and 100K cache hits free every month.

Pricing

Per-token rates, expressed in USD per 1M tokens. Verified Oct 22, 2026.

Input $0.250/M
Output $1.50/M
Blended (3:1 in:out) $0.563/M
Cross-model comparable price
Cached input $0.0250/M
Batch input $0.125/M
Batch output $0.750/M

Context & limits

Context window
1,048,576 tokens
Max input
1,048,576 tokens
Max output
65,536 tokens
Modalities
vision, audio_in, pdf, video, text

Capabilities

  • Function calling ✓ supported
  • Parallel tool calls ✓ supported
  • Vision input ✓ supported
  • Audio input ✓ supported
  • Audio output — not advertised
  • PDF input ✓ supported
  • Streaming ✓ supported
  • Structured output ✓ supported
  • Prompt caching ✓ supported
  • Reasoning ✓ supported

Strengths & caveats — when to pick Gemini 3.1 Flash Lite

Data-driven from Gemini 3.1 Flash Lite's percentile within chat peers.

Where it's strong

  • +long-context tasks — context window in the top 2% of peers
  • +audio input — only 3% of chat models on Future AGI advertise this
  • +parallel tool calls — only 18% of chat models on Future AGI advertise this
  • +PDF input — only 18% of chat models on Future AGI advertise this

Watch out for

  • !scheduled deprecation on May 7, 2027 (184 days)

Benchmark scores

Reported public benchmark numbers. Each row links to the source. Faded bar shows 6-peer average for context.

Chatbot Arena ELOgeneral· overall↑6% vs peers
Captured Aug 8, 2026
Try it

Call Gemini 3.1 Flash Lite via Agent Command Center

One OpenAI-compatible endpoint. Routing, fallback, semantic caching, guardrails, and cost tracking come along for the ride. First 100K requests + 100K cache hits free every month.

SDK
Native Future AGI client (agentcc / @agentcc/client). Per-call metadata — provider, cost, latency, cache hit, request id — is returned on x-agentcc-* response headers, so any HTTP client can read it.
# Gemini 3.1 Flash Lite via the Agent Command Center Python SDK
# pip install agentcc
import os
from agentcc import AgentCC

client = AgentCC(
    api_key=os.environ["AGENTCC_API_KEY"],   # from app.futureagi.com → Settings → API Keys
    base_url="https://gateway.futureagi.com/v1",
)

resp = client.chat.completions.create(
    model="vertex-ai/gemini-3-1-flash-lite",
    messages=[{"role": "user", "content": "Hello, Gemini 3.1 Flash Lite!"}],
)

print(resp.choices[0].message.content)
print(f"Tokens: {resp.usage.total_tokens}")

# Per-call gateway metadata is returned on x-agentcc-* response headers.
# When you need it programmatically, use .with_raw_response to get them:
raw = client.chat.completions.with_raw_response.create(
    model="vertex-ai/gemini-3-1-flash-lite",
    messages=[{"role": "user", "content": "Same call, but I want the headers."}],
)
print("Provider:", raw.headers.get("x-agentcc-provider"))
print("Latency:", raw.headers.get("x-agentcc-latency-ms"), "ms")
print("Cost:   ", raw.headers.get("x-agentcc-cost"), "USD")
print("Cache:  ", raw.headers.get("x-agentcc-cache"))
Set AGENTCC_API_KEY with a key fromapp.futureagi.com.Gateway docs ↗
Advanced: fallback + cache config (YAML) ▸
strategy: cost-optimized
targets:
  - model: gemini-3-1-flash-lite
    provider: vertex-ai
    weight: 80
fallbacks:
  - model: claude-opus-5
    provider: azure-ai-foundry
  - model: gemini-3-7-flash
    provider: google
guardrails: [pii, prompt-injection, secrets]
cache: { exact: true, semantic: true }

Same model on other providers

gemini-3-1-flash-lite is also available via 1 other route. Pricing, regions, and capabilities can differ — compare before routing production traffic.

ProviderInput / 1MOutput / 1MVerified
Google AI$0.250/M$1.50/MOct 22, 2026

Explore similar models

Grouped by Chatbot Arena tier (Gemini 3.1 Flash Lite sits at 1432 ELO). Side-by-side links appear where a comparison page exists; otherwise, open the model profile.

FAQ

How much does Gemini 3.1 Flash Lite cost?

Input is priced at $0.250 per 1M tokens and output at $1.50 per 1M tokens (Google Vertex AI, last verified Oct 22, 2026).

What is the context window of Gemini 3.1 Flash Lite?

Gemini 3.1 Flash Lite supports a 1,048,576-token context window with up to 65,536 output tokens.

Does Gemini 3.1 Flash Lite support function calling?

Yes — Gemini 3.1 Flash Lite supports function (tool) calling, including parallel tool calls.

Is Gemini 3.1 Flash Lite good for production?

Gemini 3.1 Flash Lite is well-suited for long-context tasks — context window in the top 2% of peers and audio input — only 3% of chat models on Future AGI advertise this. Consider alternatives if you need scheduled deprecation on May 7, 2027 (184 days).

How can I route to Gemini 3.1 Flash Lite with fallback?

Use Agent Command Center: a single OpenAI-compatible endpoint that supports cost-optimized routing, latency-aware retries, model fallback, and shadow traffic. Configure once, swap models without app changes.

Useful links for Gemini 3.1 Flash Lite

Official sources, independent benchmarks, and pricing aggregators — no random search-engine guesses.