Dated comparison: 3 September 2026
| Metric | Gemini 3.8 Flash | Claude Opus 5 |
|---|---|---|
| AA Intelligence Index | 59 | 63 |
| Terminal-Bench 2.1 | 89.4% | 89.1% |
| Terminal-Bench 4.0 | 19.1% | 51.8% |
| OSWorld 2.0 | 59.0% | 75.4% |
| GDPval-AA v2 | 1545 Elo | 1824 Elo |
| Vals Finance Agent v2 | 61.4% | 58.6% |
| Harvey Legal Agent | 10.0% | 6.7% |
| LVBench (agentic) | 87.8% | 75.4% |
| HLE-Verified | 54.9% | 54.4% |
| DeepSWE v1.1 | 73.7% | 74.0% |
| Standard input / 1M | $0.75 | ~$5 |
| Standard output / 1M | $3.75 | ~$30 |
| Time to first token | 13.39 s | Lower |
| Weights | Closed | Closed |
How to choose
Pick Gemini 3.8 Flash for finance agents, legal document review, long video, batch processing and cost-bound coding tasks. Pick Opus 5 for open-ended long-horizon agents, computer use and broad professional work. The four losses above 15 points are the place Gemini's head-to-head positioning is weakest.
Ready to test the workflow?
Create account & add creditsComparison limits
These figures are transcribed from the 3 September 2026 research snapshot. Opus 5's per-token prices are approximate ranges reported by third-party aggregators; verify against your billing portal before quoting.
Frequently asked questions
How current is this information?
This page uses the verified research snapshot dated 3 September 2026. Verify current access, pricing and benchmark figures before a production decision.
Put Gemini 3.8 Flash to work
Fund a controlled evaluation, send a reference frame or document, and measure quality and cost on your own workload.