Effort: high · High
googleReleased Sep 30, 2026
Google: Gemini 4 Argon
Long decode can use the unified 1M-token budget for output; reads video, audio and PDFs, with implicit-only caching.
Source description
Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...
Context window1M
Max output65.5K
Provider endpoints8
Benchmark results19
Measured performance
All leaderboards ↗Benchmark results
Effort: high · High
GDPval-AA ↗2.1
1,623.99Effort: high · High
AutomationBench-AA ↗AA Index 4.3.2 collection
77.51%Effort: high · High
Effort: high · High · Harness unreported
SciCode ↗AA Index 4.3.2 collection
61.81%AA · Under reviewEffort: high · High
Humanity’s Last Exam ↗AA Index 4.3.2 collection
57.09%Effort: high · High
GDP.pdf ↗AA Index 4.3.2 collection
21.8%Effort: high · High
CritPt ↗AA Index 4.3.2 collection
27.14%AA · Under reviewEffort: high · High
AA-Omniscience ↗AA Index 4.3.2 collection
42.35Effort: high · High
AA-LCR ↗1.1
79.67%Effort: high · High
LiveBench Overall ↗2.0 · 2026-10-14 (dataset 2026-09-02)
84.1%Effort: high · single run · equal mean of 7 suites
LiveBench Design ↗2.0 · 2026-10-14 (dataset 2026-09-02)
73.7%Effort: high · single run
LiveBench Research ↗2.0 · 2026-10-14 (dataset 2026-09-02)
83.1%Effort: high · single run
LiveBench Browser Use ↗2.0 · 2026-10-14 (dataset 2026-09-02)
92%Effort: high · single run
LiveBench Coding ↗2.0 · 2026-10-14 (dataset 2026-09-02)
89.7%Effort: high · single run
LiveBench Agentic ↗2.0 · 2026-10-14 (dataset 2026-09-02)
69.9%Effort: high · single run
LiveBench Data Analysis ↗2.0 · 2026-10-14 (dataset 2026-09-02)
85.2%Effort: high · single run
LiveBench Reasoning ↗2.0 · 2026-10-14 (dataset 2026-09-02)
95.1%Effort: high · single run
Access & cost
8 quotesPrices across providers
Your token mix
USD · listed token ratesOpenRouter
Catalog quote- Input
- $2.00
- Output
- $10.00
- Cached input
- $0.10
Experiential
Standard list rate
- Input
- $2.00
- Output
- $10.00
- Cached input
- $0.10
Catalog list rate. Plan allowances and temporary discounts are separate.
- Input
- $1.00
- Output
- $5.00
- Cached input
- $0.05
Cache write $0.09375/M tokens · Audio input $1/M audio tokens · Cached audio input $0.05/M audio tokens · Audio tokens use separate rates from text input · Web search $0.014/search · Source discount 50%; rates as published
- Input
- $2.00
- Output
- $10.00
- Cached input
- $0.10
Cache write $0.1875/M tokens · Audio input $2/M audio tokens · Cached audio input $0.1/M audio tokens · Audio tokens use separate rates from text input · Web search $0.014/search · Source discount 50%; rates as published
- Input
- $2.00
- Output
- $10.00
- Cached input
- $0.10
Cache write $0.1875/M tokens · Audio input $2/M audio tokens · Cached audio input $0.1/M audio tokens · Audio tokens use separate rates from text input · Web search $0.014/search · Source discount 50%; rates as published
- Input
- $2.20
- Output
- $11.00
- Cached input
- $0.11
Cache write $0.1875/M tokens · Audio input $2.2/M audio tokens · Cached audio input $0.11/M audio tokens · Audio tokens use separate rates from text input · Web search $0.014/search · Source discount 50%; rates as published
- Input
- $3.60
- Output
- $18.00
- Cached input
- $0.18
Cache write $0.3375/M tokens · Audio input $3.6/M audio tokens · Cached audio input $0.18/M audio tokens · Audio tokens use separate rates from text input · Web search $0.014/search · Source discount 50%; rates as published
- Input
- $3.60
- Output
- $18.00
- Cached input
- $0.18
Cache write $0.3375/M tokens · Audio input $3.6/M audio tokens · Cached audio input $0.18/M audio tokens · Audio tokens use separate rates from text input · Web search $0.014/search · Source discount 50%; rates as published
Offers for this model
1 offerModel specifications
- Developer
- Modality
- Language
- Context
- 1,048,576 tokens
- Maximum output
- 65,536 tokens
- Weights
- Not reported
- Released
- Sep 30, 2026
- Provider listed
- Oct 13, 2026
- First discovered
- Oct 16, 2026
Exact identities
- openrouter
google/gemini-4-argon - experiential
gemini-4-argon - artificial-analysis
d366dfa1-d940-4788-aa73-d298c45699b3 - livebench
gemini-4-argon
Sources
Updated Oct 18, 2026
More from google
View all ↗googleListed 4 days ago
Google: Gemini 4 Argon (batch)
Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep…
1M contextImage inputAudio inputNo reported score
No reported score
googleListed 22 days ago
Google: Gemini 3.5 Transcribe
Gemini 3.5 Transcribe is a speech-to-text model from Google. It is suited for synchronous transcription that needs word-level timestamps or speaker diarization, with support for up to eight speakers. Audio...
98.3K contextAudio inputNo reported score
No reported score
googleListed Sep 2
Google: Gemini 3.8 Flash
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
1M contextImage inputAudio inputNo reported score
No reported score
$0.375in$1.875out/ 1M tokens · USDGoogle AI Studio · via OpenRouter · 9 endpointsgoogle-ai-studio/flex · tier=flex
googleListed Sep 2
Google: Gemini 3.8 Flash (batch)
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
1M contextImage inputAudio inputNo reported score
No reported score