AI Model Release Tracker

Gemini 3.0 Flash

Released

Gemini 3.0 Flash is an AI model released by Google on Wednesday, Dec 17 2025, 29 days after Gemini 3.0 Pro. Benchmark results (shown below) cover BullshitBench v2, ProgramBench, SWE-Bench Pro, SWE-Bench Verified, Terminal-Bench 2.1, MCP Atlas, and 11 more.

Available from

Google AI Studio—flex$0.25$1.501M
Google AI Studio——$0.50$3.001M
Googleglobal—$0.50$3.001M
Google AI Studio—priority$0.90$5.401M
Googleglobalpriority$0.90$5.401M
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Checked October 16, 2026Third-party rates from openrouter.ai, not the lab’s own list price.

Benchmarks

Coding

ProgramBenchProgram reconstruction — The AI receives a working program and its documentation, then builds a replacement from scratch without the original source code, internet access or decompilation. The score is the percentage of 200 programs that pass every behavioral test. We record each model's best published mini-SWE-agent result, including higher reasoning efforts where available. Partial test-pass rates and almost-solved programs do not count toward this score. Equal scores share a rank here; the official board also uses partial progress to break ties. Higher is better.via ProgramBench
0%
#6 of 20
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better.
49.6%
#22 of 24
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.
78%
#18 of 59

Terminal & CLI

Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better.
58%
#31 of 35

Agentic & tool use

MCP AtlasMulti-step tool use — Can the AI chain together many tools and steps to complete one bigger task, rather than doing just a single thing? Higher is better.
62%
#12 of 13
ToolathlonGeneral tool use — Tests how well the AI uses everyday real-world tools and apps to get things done. Higher is better.
49.4%
#4 of 5
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
65.1%
#22 of 29

Reasoning & science

Humanity's Last ExamMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better.
33.7%no tools
#15 of 23
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better.
33.6%
#33 of 37
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better.
90.4%
#16 of 61

Knowledge work

GDPval-AAKnowledge work — Measures how well the AI does economically valuable knowledge work, judged against human experts. Shown as a rating (like a chess Elo) — higher is better.
1204
#9 of 9

Finance

Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better.
42.6%
#12 of 12

Multimodal

CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better.
80.3%
#13 of 16
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better.
81.2%
#5 of 12
Blueprint-Bench 2Spatial reasoning — Can the AI reason about space and layout — for example, understanding a floor plan or blueprint? Higher is better.
0%
#6 of 6

Long context

MRCR v2 (8-needle)Long context — Tests whether the AI can find specific details buried inside a very long document (around 128k tokens — roughly a long book). Higher is better.
67.2%128k average
22.1%1M pointwise
128k average
#6 of 7
1M pointwise
#4 of 4

Robustness

BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better.
10%
#86 of 97

Source: ProgramBench, retrieved 11 September 2026. Other sources are identified on the linked benchmark pages.

Compare Gemini 3.0 Flash with

Gemini 3.0 Flash

Suggested comparisons

Gemini 3.0 FlashvsGemini 4 ArgonGemini 3.0 FlashvsGPT-6.1 SolGemini 3.0 FlashvsClaude Haiku 5.5Gemini 3.0 FlashvsMuse Spark OpenGemini 3.0 FlashvsGrok 4.7Gemini 3.0 FlashvsDeepSeek-V4.1-FlashGemini 3.0 FlashvsMistral Large 4Gemini 3.0 FlashvsKimi K3Gemini 3.0 FlashvsGLM-5.3-FlashGemini 3.0 FlashvsQwen3.8-Max-0902Gemini 3.0 FlashvsNemotron 3.5 Super VL

Frequently asked questions

Gemini 3.0 Flash was released by Google on Wednesday, Dec 17 2025.

All Google releases

32 tracked