Claude Opus 4.6 vs Gemini 4 Argon
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Opus 4.6—
Gemini 4 Argon77.5
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 4.667
Gemini 4 Argon79.7
EBR Bench EBR Bench: Agents learn to play Earthborne Rangers across repeated playthroughs
Opus 4.6 · max12.7
Gemini 4 Argon64.8
Furniture Assembly Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step
Opus 4.6 · max28.3
Gemini 4 Argon72.5
AutomationBench · Zapier run AutomationBench · Zapier run: End-to-end completion of business workflows using real SaaS applications
Opus 4.6—
Gemini 4 Argon51.3
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 4.6—
Gemini 4 Argon56.2
MCP-Atlas Public MCPAtlas Public: MCPAtlas Public as reported in this launch evaluation; the post is the authority for the recorded task and grading scope
Opus 4.6 · max79
Gemini 4 Argon—
MCP-Atlas · full 1000 MCP-Atlas · full 1000: Tool orchestration across MCP servers on the stated task split
Opus 4.6 · max76.8
Gemini 4 Argon—
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 4.6 · max27.3
Gemini 4 Argon—
AA Analyst Agent AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings
Opus 4.6—
Gemini 4 Argon48.8
APEX-Agents APEX-Agents: Agents complete long-horizon professional-services tasks across realistic applications
Opus 4.6 · max33
Gemini 4 Argon—
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 4.6—
Gemini 4 Argon57.1
Terminal-Bench Hard (version unstated) Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal
Opus 4.648.5
Gemini 4 Argon—
Terminal-Bench Science 0.1 Terminal-Bench Science 0.1: Agents complete expert-authored research workflows across five scientific domains in a terminal
Opus 4.6—
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 4.6—
Gemini 4 Argon1,485.8
DeepSWE 1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition
Opus 4.6 · max31.9
Gemini 4 Argon—
FrontierCode 1.1 FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements
Opus 4.6—
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 4.6—
Gemini 4 Argon3.9
SWE-bench Pro (private) SWE-bench Pro (private): Agents resolve long-horizon professional repository tasks; public and private splits are separate
Opus 4.647.1
Gemini 4 Argon—
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 4.6—
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 4.6—
Gemini 4 Argon92.2
| Benchmark | Opus 4.6 | Gemini 4 Argon |
|---|---|---|
| AutomationBench-AA | — | 77.5 |
| AA-LCR (long context) 1.1 | 67 | 79.7 |
| EBR Bench | 12.7 · max | 64.8 |
| Furniture Assembly | 28.3 · max | 72.5 |
| AutomationBench · Zapier run | — | 51.3 |
| GDPval-AA (work tasks) 2.1 | — | 56.2 |
| MCP-Atlas Public | 79 · max | — |
| MCP-Atlas · full 1000 | 76.8 · max | — |
| τ³-bench · banking · 1.0.1 | 27.3 · max | — |
| AA Analyst Agent | — | 48.8 |
| APEX-Agents | 33 · max | — |
| Terminal-Bench 4.0 | — | 57.1 |
| Terminal-Bench Hard (version unstated) | 48.5 | — |
| Terminal-Bench Science 0.1 | — | 52.9 |
| AA-Briefcase 1.1 · Elo | — | 1,485.8 |
| DeepSWE 1.1 | 31.9 · max | — |
| FrontierCode 1.1 | — | 43.5 |
| Harvey LAB-AA 1.1 | — | 3.9 |
| SWE-bench Pro (private) | 47.1 | — |
| SWE-bench Pro (v2 full) | — | 99.2 |
| SWE-bench Pro (v2 hard) | — | 92.2 |
Specs
| Spec | Opus 4.6 | Gemini 4 Argon |
|---|---|---|
| Input, per 1M | $5 | !$2 |
| Output, per 1M | $25 | !$10 |
| Cost per task? | — | $1.99 |
| Speed, tok/s | 45 | 84 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Feb 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 4.6 → 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 4.6
Nothing known.