Cheaper overall
Gemini 4 Argon
60% lower blended rate
Higher benchmark scores
Gemini 4 Argon
leads on 11 of 16 figures
Bigger context window
Gemini 4 Argon
1,048,576 vs 1,000,000 tokens
Cheaper cached input
Gemini 4 Argon
$0.10 vs $0.50, 80% less
Where each one wins
Gemini 4 Argon
- Cheaper input tokens $2.00 vs $5.00, 60% less
- Cheaper output tokens $10.00 vs $25.00, 60% less
- Cheaper cached input $0.10 vs $0.50, 80% less
- Cheaper on all four workloads
- Larger context window 1,048,576 tokens
- Higher overall intelligence (Intelligence Index) 52.6 vs 50.8
- Higher expert-exam performance (Humanity's Last Exam) 57.1 vs 54.9
- Higher scientific coding (SciCode) 61.8 vs 56.4
- Ahead on 8 more benchmark figures
Claude Opus 5
- Longer maximum output 128,000 tokens
- Higher long-context reasoning (AA-LCR) 82 vs 79.7
- Higher physics research reasoning (CritPt) 29.1 vs 27.1
- Higher real-world task performance (GDPval) 61.1 vs 56.2
- Ahead on 2 more benchmark figures
- Offers Batch pricing not listed for Gemini 4 Argon
What four workloads cost
One run, and the same run a thousand times.
Gemini 4 Argon is cheaper on all four workloads, by much the same margin each time — about 2.6 times, from $0.0070 against $0.0175 on a chat turn to $0.8300 against $2.08 on a long document.
| Workload | Gemini 4 Argon | Claude Opus 5 | ×1,000 | Difference |
|---|---|---|---|---|
| Chat turn 1,000 in / 500 out | $0.0070 | $0.0175 | $7.00 / $17.50 | Gemini 4 Argon is 60% cheaper |
| RAG answer 10,000 in / 800 out | $0.0280 | $0.0700 | $28.00 / $70.00 | Gemini 4 Argon is 60% cheaper |
| Agent loop 32,000 in / 700 out × 20 calls | $0.2800 | $0.8500 | $280.00 / $850.00 | Gemini 4 Argon is 67% cheaper |
| Long document 400,000 in / 3,000 out | $0.8300 | $2.08 | $830.00 / $2,075.00 | Gemini 4 Argon is 60% cheaper |
Price per 1M tokens
Neither model changes its rate with prompt length, so these rates apply to every request.
| Context band | Price | Gemini 4 Argon | Claude Opus 5 | Difference |
|---|---|---|---|---|
| Any prompt length | Input | $2.00 | $5.00 | Gemini 4 Argon is 60% cheaper |
| Any prompt length | Cached input | $0.10 | $0.50 | Gemini 4 Argon is 80% cheaper |
| Any prompt length | Output | $10.00 | $25.00 | Gemini 4 Argon is 60% cheaper |
Blended: Gemini 4 Argon $4.00, Claude Opus 5 $10.00 per 1M — one rate at 3:1 input to output, for comparing two models at a glance.
Gemini 4 Argon priced by Perplexity, Claude Opus 5 by Anthropic. Standard tier, pay-as-you-go.
Other pricing tiers
Only Claude Opus 5 lists Batch, at up to 50% off its own standard rate; Gemini 4 Argon does not offer it.
Every tier either model sells is here, so this is the whole price sheet. A saving is measured against that model's own standard rate — how we read tiers.
| Tier | Gemini 4 Argon | Claude Opus 5 | Cheaper |
|---|---|---|---|
| Batch only on Claude Opus 5 | Not offered | $2.50 in / $12.50 out 50% off Standard | — |
Benchmarks
They share 17 figures: Gemini 4 Argon leads on 11, Claude Opus 5 on 5. The widest gap is 239 points, on Blender3d Elo.
Overall intelligence
Intelligence Index — Overall intelligence across reasoning, knowledge & math evals
Gemini 4 Argon ahead by 1.8
Expert-exam performance
Humanity's Last Exam — Expert-level questions across many academic domains
Gemini 4 Argon ahead by 2.2
Long-context reasoning
AA-LCR — Long-context reasoning across large inputs
Claude Opus 5 ahead by 2.3
Scientific coding
SciCode — Research-level scientific coding tasks
Gemini 4 Argon ahead by 5.4
Physics research reasoning
CritPt — Unpublished physics research reasoning problems
Claude Opus 5 ahead by 2
Real-world task performance
GDPval — Economically valuable, real-world knowledge work
Claude Opus 5 ahead by 4.9
Factual accuracy
Omniscience Accuracy — Breadth of factual knowledge across domains
Claude Opus 5 ahead by 11
Answer reliability
Omniscience Non-Hallucination — How reliably the model avoids fabricated answers
Gemini 4 Argon ahead by 44.4
Blender3d Elo
Design Arena head-to-head rating in the blender3d category
Gemini 4 Argon ahead by 239
Spreadsheet Elo
Design Arena head-to-head rating in the spreadsheet category
Gemini 4 Argon ahead by 33
3d Elo
Design Arena head-to-head rating in the 3d category
Gemini 4 Argon ahead by 52
Codecategories Elo
Design Arena head-to-head rating in the codecategories category
Gemini 4 Argon ahead by 7
Dataviz Elo
Design Arena head-to-head rating in the dataviz category
Level
Gamedev Elo
Design Arena head-to-head rating in the gamedev category
Gemini 4 Argon ahead by 31
Svg Elo
Design Arena head-to-head rating in the svg category
Gemini 4 Argon ahead by 23
Uicomponent Elo
Design Arena head-to-head rating in the uicomponent category
Claude Opus 5 ahead by 13
Website Elo
Design Arena head-to-head rating in the website category
Gemini 4 Argon ahead by 7
Each figure is that model's best published run, with the effort level named beside it — where these scores come from.
Specifications
Gemini 4 Argon holds 48,576 more input tokens in one request. Gemini 4 Argon also takes audio and video. Both list the same capabilities, so nothing here separates them.
| Specification | Gemini 4 Argon | Claude Opus 5 |
|---|---|---|
| Context window | 1,048,576 tokens | 1,000,000 tokens |
| Max output | 65,536 tokens | 128,000 tokens |
| Takes in | Text, audio, image, video, file | Text, image, file |
| Puts out | Text | Text |
| Knowledge cutoff | 31 Mar 2026 | — |
| Released | 13 Oct 2026 | 24 Jul 2026 |
| Status | Active | Active |
| Sold by | Perplexity | Anthropic, Perplexity |
| Tool use | Yes | Yes |
| Structured outputs | Yes | Yes |
| Web search | Not listed | Not listed |
| Prompt caching | Yes | Yes |
| Code execution | Yes | Yes |
| Computer use | Yes | Yes |
| Open weights | Not listed | Not listed |
| Reasoning | Yes | Yes |
FAQs
- Is Gemini 4 Argon cheaper than Claude Opus 5?
- Yes. A 1,000-token prompt with a 500-token reply costs $0.0070 on Gemini 4 Argon and $0.0175 on Claude Opus 5, and the same model is cheaper on every workload on this page.
- How much do Gemini 4 Argon and Claude Opus 5 cost per 1M tokens?
- Gemini 4 Argon costs $2.00 for input and $10.00 for output. Claude Opus 5 costs $5.00 and $25.00. Standard pay-as-you-go rates, as of 17 Oct 2026.
- Which is better, Gemini 4 Argon or Claude Opus 5?
- On published benchmarks, Gemini 4 Argon. It leads on 11 of the 17 figures both models report, including overall intelligence (Intelligence Index), where it scores 52.6 against 50.8.
- Does Gemini 4 Argon or Claude Opus 5 offer batch pricing?
- Claude Opus 5 only. Its batch input costs $2.50, 50% off its standard rate, and Gemini 4 Argon lists no batch tier at all.
- Does Gemini 4 Argon or Claude Opus 5 have a bigger context window?
- Gemini 4 Argon, at 1,048,576 tokens against 1,000,000. That is 48,576 more input tokens in a single request.