LLM compare · detail

Claude Sonnet 4.6

Provider: Anthropic

Model type: specialist Budget: mid-budget Use it for: coding Use it for: AI agents Use it for: company knowledge

coding pick backed by current AA.

Verdict „Claude Sonnet 4.6 has IQ 51.7 and input $3/M and DeepSWE pass@1 31.8 % in the sources. Consider it for coding, AI agents; for self-hosted stack, mass-volume, run a second benchmark before rollout."

Snapshot summary

Anthropic is the US AI company behind Claude, strongest around long context, coding workflows, agent control and safer model behavior. Claude Sonnet 4.6 is the working Claude tier below Opus. It usually fits everyday coding and agent work when Opus gets too expensive.

In this compare snapshot it is tagged mainly for coding, agents, rag, with price tier mid and context window není v ověřených datech. The negative side matters too: watch self-hosted, mass-volume. 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

$3/M input artificial_analysis:claude-sonnet-4-6-adaptive

$15/M output artificial_analysis:claude-sonnet-4-6-adaptive

51.7 intelligence index artificial_analysis:claude-sonnet-4-6-adaptive

31.8% DeepSWE pass@1 deepswe:claude-sonnet-4-6

61.9% DeepSWE pass@4 · high deepswe:claude-sonnet-4-6

What this means

✓ Good for: coding · agents · rag

✗ Not for: self-hosted · mass-volume

Source and provenance

Data source: artificial_analysis
Fetched: 2026-06-14T04:00:40Z
Status per metric: price: artificial_analysis:claude-sonnet-4-6-adaptive; IQ: artificial_analysis:claude-sonnet-4-6-adaptive; Elo: curated_seed; Aider: curated_seed; DeepSWE: deepswe:claude-sonnet-4-6

This page shows only verified data from the current snapshot.