LLM compare · detail

Gemini 3.1 Pro Preview

Provider: Google

Model type: market frontier Budget: mid-budget Use it for: company knowledge Use it for: vision and multimodal Use it for: multilingual content

rag pick backed by current AA.

Verdict „Gemini 3.1 Pro Preview has IQ 57.2 and input $2/M and DeepSWE pass@1 9.7 % in the sources. Consider it for company knowledge, vision and multimodal; for real-time-latency, self-hosted stack, run a second benchmark before rollout."

Snapshot summary

Google positions Gemini as a model layer for its cloud, workspace and multimodal ecosystem, so it covers text, large context, video, audio and vision. Gemini 3.1 Pro Preview is the larger Gemini option for big inputs, multimodality and tasks where the Google ecosystem is an advantage, not baggage.

In this compare snapshot it is tagged mainly for rag, vision, multilingual, with price tier mid and context window není v ověřených datech. The negative side matters too: watch real-time-latency, self-hosted. Use the numbers below as the hard boundary, because this profile explains the model, it does not replace benchmark provenance.

Figures below, fetched from source artificial_analysis (fetched 2026-06-14T04:00:40Z).

Snapshot figures

$2/M input artificial_analysis:gemini-3-1-pro-preview

$12/M output artificial_analysis:gemini-3-1-pro-preview

57.2 intelligence index artificial_analysis:gemini-3-1-pro-preview

9.7% DeepSWE pass@1 deepswe:gemini-3-1-pro-preview

24.8% DeepSWE pass@4 deepswe:gemini-3-1-pro-preview

What this means

✓ Good for: rag · vision · multilingual

✗ Not for: real-time-latency · self-hosted

Source and provenance

Data source: artificial_analysis
Fetched: 2026-06-14T04:00:40Z
Status per metric: price: artificial_analysis:gemini-3-1-pro-preview; IQ: artificial_analysis:gemini-3-1-pro-preview; Elo: curated_seed; Aider: curated_seed; DeepSWE: deepswe:gemini-3-1-pro-preview

This page shows only verified data from the current snapshot.