Google Gemini model

Gemini 2.5 Flash

Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.

Provider
Google Gemini
Availability
Available
Model ID
gemini-2.5-flash
Snapshot
2026-10-07
Evaluation Verdict

Gemini 2.5 Flash is a high-speed, 1M context utility workhorse (180 TPS, $0.075/M input) scoring 58.2% on SWE-bench Verified, offering unbeatable economics for background agent tasks.

Gemini 2.5 Flash Benchmarks & Agentic Accuracy

Verified primary research & developer preference metrics

Primary Sources Verified
SWE-bench Verified
58.2%
StandardGoogle DeepMind Report ↗
LiveCodeBench
42.1%
StandardLiveCodeBench ↗
LMSYS Coding Elo
1330Elo
Coding ArenaLMSYS Arena ↗
AIME 2024 / 2025
68%
StandardGoogle DeepMind Report ↗
Pricing ($ / 1M tokens)
$0.075/ $0.3
$0.01875 cached readStandard API Pricing

What This Means in Real Software Development

  • Blazing 180 TPS generation speed with near-instant time-to-first-token.
  • Scores 58.2% on SWE-bench Verified at ultra-low $0.075/M input pricing.
  • 1M token context window fits full build logs and stack traces with ease.
  • Highly stable for repetitive subagent execution loops and test validation.

Swarm Multi-Agent Topology

Tier 3: High-Velocity Subagent EngineRole: Finder Subagent

Best Paired With: Claude 3.7 Sonnet as Lead Planner

High throughput and low pricing make it the premier engine for concurrent Coder and Finder jobs.

Git Worktree Isolation Advantage:

Executes parallel bug fixes across isolated worktrees without file contention.

What Developers & Practitioners Are Saying

Real-world field observations, production strengths, and developer consensus

Field Consensus

"High-volume workhorse offering 1M context at ultra-budget rates ($0.075/M input)."

Verified Real-World Strengths

  • Near-zero cost for processing massive logs and documentation dumps
  • Fast 180 TPS generation speed
  • Strong multimodal support

Production Caveats & Gaps

  • Can lose track of complex negative instructions in long prompts
Benchmark Reality Check

What the benchmark tests: Tests low-cost context ingestion and rapid summarization.

Production caveat: Not recommended for deep AST refactoring; best deployed for high-volume context analysis.

Primary Sources & Methodology Citations

Gemini 2.5 Flash model overview

Gemini 2.5 Flash is listed in the public Swarm catalog through Google Gemini. This page summarizes the verified catalog fields used by Swarm without claiming real-time access for every provider account.

Provider
Google Gemini
Family
Gemini
Context window
1.05M tokens
Thinking support
off
Availability status
Available
Catalog ID
google/gemini-2.5-flash
  • Audio input
  • Image input
  • Text input
  • Video input
  • Text output
  • Reasoning
  • Batch

Run Gemini 2.5 Flash in Swarm

Verified CLI workflow & multi-agent architecture

Quickstart Guide →
bash — swarm run
$swarm ctl model set --provider google --model google/gemini-2.5-flash

# Or switch interactively inside the terminal TUI:

>/models

Subagent Role Matrix

Assign Gemini 2.5 Flash to specialized system roles in ~/.swarm/config.json:

  • Primary / Planner: Orchestrates full task DAGs and multi-stage pipelines.
  • Coder Subagent: Generates code and unit tests inside private Git worktrees.
  • Finder Subagent: Performs read-only code search, symbol discovery, and mapping.
← All Google Gemini models

Public catalog snapshot. Catalog availability is not a real-time provider status or an access guarantee.