Claude Opus 5 vs Gemini 4 Argon
Different efforts: medium · high · Compare all at high
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Opus 554.3
Gemini 4 Argon77.5
GDPval-AA (work tasks) 2.1 GDPval-AA (work tasks) 2.1: Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work
Opus 549.6
Gemini 4 Argon56.2
AA Analyst Agent AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings
Opus 5 · max53.8
Gemini 4 Argon48.8
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Opus 534.3
Gemini 4 Argon57.1
Terminal-Bench Science 0.1 Terminal-Bench Science 0.1: Agents complete expert-authored research workflows across five scientific domains in a terminal
Opus 5 · max28.6
Gemini 4 Argon52.9
AA-Briefcase 1.1 · Elo AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating
Opus 51,434.9
Gemini 4 Argon1,485.8
AA-LCR (long context) 1.1 AA-LCR (long context) 1.1: Questions that need reading and connecting facts across very long documents (around 100K tokens)
Opus 582
Gemini 4 Argon79.7
SWE-bench Pro (v2 full) SWE-bench Pro (v2 full): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Opus 5 · xhigh99.4
Gemini 4 Argon99.2
SWE-bench Pro (v2 hard) SWE-bench Pro (v2 hard): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Opus 5 · xhigh98
Gemini 4 Argon92.2
EBR Bench EBR Bench: Agents learn to play Earthborne Rangers across repeated playthroughs
Opus 5 · max45.7
Gemini 4 Argon64.8
Furniture Assembly Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step
Opus 5 · max60.8
Gemini 4 Argon72.5
AutomationBench · Zapier run AutomationBench · Zapier run: End-to-end completion of business workflows using real SaaS applications
Opus 5—
Gemini 4 Argon51.3
τ³-bench · banking · 1.0.1 τ-bench official · 3-banking-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Opus 5 · max48.7
Gemini 4 Argon—
τ³-Banking · AA τ³-Banking · AA: Agents retrieve banking knowledge and execute tools to resolve customer-support workflows
Opus 538.6
Gemini 4 Argon—
DeepSWE 1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition
Opus 568.9
Gemini 4 Argon—
EnterpriseOps-Gym-AA EnterpriseOps-Gym-AA: Agents complete stateful enterprise workflows in eight domains; final database state determines success
Opus 5 · max47.5
Gemini 4 Argon—
FrontierCode 1.1 FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements
Opus 5—
Gemini 4 Argon43.5
Harvey LAB-AA 1.1 Harvey LAB-AA 1.1: Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit
Opus 5—
Gemini 4 Argon3.9
| Benchmark | Opus 5 | Gemini 4 Argon |
|---|---|---|
| AutomationBench-AA | 54.3 | 77.5 |
| GDPval-AA (work tasks) 2.1 | 49.6 | 56.2 |
| AA Analyst Agent | 53.8 · max | 48.8 |
| Terminal-Bench 4.0 | 34.3 | 57.1 |
| Terminal-Bench Science 0.1 | 28.6 · max | 52.9 |
| AA-Briefcase 1.1 · Elo | 1,434.9 | 1,485.8 |
| AA-LCR (long context) 1.1 | 82 | 79.7 |
| SWE-bench Pro (v2 full) | 99.4 · xhigh | 99.2 |
| SWE-bench Pro (v2 hard) | 98 · xhigh | 92.2 |
| EBR Bench | 45.7 · max | 64.8 |
| Furniture Assembly | 60.8 · max | 72.5 |
| AutomationBench · Zapier run | — | 51.3 |
| τ³-bench · banking · 1.0.1 | 48.7 · max | — |
| τ³-Banking · AA | 38.6 | — |
| DeepSWE 1.1 | 68.9 | — |
| EnterpriseOps-Gym-AA | 47.5 · max | — |
| FrontierCode 1.1 | — | 43.5 |
| Harvey LAB-AA 1.1 | — | 3.9 |
Specs
| Spec | Opus 5 | Gemini 4 Argon |
|---|---|---|
| Input, per 1M | $5 | !$2 |
| Output, per 1M | $25 | !$10 |
| Cost per task? | $2.19 | $1.99 |
| Speed, tok/s | 55 | 84 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Jul 2026 | Oct 2026 |
| See images | ✓ | ✓ |
| Read PDFs | Partial | ✓ |
| Hear audio | — | ✓ |
| Watch video | — | ✓ |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ✓ |
| Run code | ? | ✓ |
| Use a computer | ? | ? |
What breaks when you switch
Opus 5 → Gemini 4 Argon
Synchronous output stops at 65,536›
With background false, output limits above 65,536 tokens return HTTP 400; stay at or below that limit.
Sampling parameters are ignored›
temperature, top_p and top_k are ignored; remove them from requests.
No explicit caching›
Explicit caching is unavailable; omit explicit cache requests.
Gemini 4 Argon → Opus 5
Nothing known.