Claude Sonnet 4.6 — Spezifikationen
Compiled by Mustafa Ihsan from the vendor’s published documentation · Last updated
| Entwickler | Anthropic |
|---|---|
| Typ | LLM |
| Modality | Text, Vision → Text |
| Parameter | Nicht offengelegt |
| Kontextfenster | 1 Mio. |
| Maximale Ausgabe | 64 K |
| Lizenz | Proprietär |
| Offene Gewichte | Nein |
| Veröffentlicht | 2025 |
| Eingabepreis | $3.00 /1M |
| Ausgabepreis | $15.00 /1M |
| API-Anbieter | Anthropic, AWS, Vertex AI, Azure |
What is Claude Sonnet 4.6?
Claude Sonnet 4.6 is Anthropic’s balance point between speed and intelligence: adaptive
thinking, a 1M-token context window, 64K maximum output, and pricing of $3 in / $15 out per
million tokens. It is the default workhorse for most production workloads, and the tier most
teams land on after discovering that Opus is overkill for the majority of their traffic and
Haiku is underpowered for the rest.
What makes Sonnet the pragmatic pick is that it gives up very little at the top of the
range while costing 40% less than Opus 4.8 on both input and output. The 1M context is
identical, adaptive thinking is identical, and the gap only becomes visible on genuinely
hard multi-step reasoning. For retrieval-augmented generation, code review, structured
extraction over long documents and customer-facing assistants, that gap rarely shows up in
output quality but always shows up on the invoice. Note that Claude Sonnet 5 now sits below
it on price — $2 / $10 during its introductory period — so a new project should compare the
two directly before committing to 4.6.
Claude Sonnet 4.6 pricing: API cost per 1M tokens
| Eingabe (pro 1 Mio. Token) | $3.00 |
|---|---|
| Ausgabe (pro 1 Mio. Token) | $15.00 |
| Verhältnis Output/Input | 5× |
| Gemischt (4:1 Input:Output) | $5.40 pro 1 Mio. Tokens |
What Claude Sonnet 4.6 costs per month
Tatsächliche monatliche Ausgaben bei einem 4:1-Input-zu-Output-Mix – dem Verhältnis, das typische Chat- oder RAG-Arbeitslasten tatsächlich erzeugen.
| Workload | Tokens/Monat | Kosten pro Monat |
|---|---|---|
| Nebenprojekt | 1 Mio. Eingabe / 0,25 Mio. Ausgabe | $6.75 |
| Kleines Team | 20 Mio. Eingabe / 5 Mio. Ausgabe | $135 |
| Produktion | 200 Mio. Eingabe / 50 Mio. Ausgabe | $1,350 |
Stellen Sie Ihre eigenen Berechnungen im KI-API-Kostenrechner.
Cheaper alternatives to Claude Sonnet 4.6
| Modell | Gewichteter Preis pro Million US-Dollar | Sie sparen |
|---|---|---|
| GLM 5.2 offenere | $2.00 | 63 % günstiger |
| Gemini 3.5 Flash | $3.00 | 44 % günstiger |
| Gemini 3.1 Pro | $4.00 | 26 % günstiger |
Häufig gestellte Fragen
How much does Claude Sonnet 4.6 cost per 1M tokens?
Claude Sonnet 4.6 costs $3.00 per 1M input tokens and $15.00 per 1M output tokens. At a typical 4:1 input-to-output mix that blends to about $5.40 per 1M tokens.
How much does Claude Sonnet 4.6 cost per month?
A small-team workload of 20M input and 5M output tokens a month costs about $135 on Claude Sonnet 4.6. A side project (1M in / 0.25M out) costs roughly $6.75.
What is a cheaper alternative to Claude Sonnet 4.6?
GLM 5.2 is the strongest cheaper option in our database at $2.00 per 1M blended — about 63% less than Claude Sonnet 4.6. It is also open-weight, so self-hosting is an option.
Can I run Claude Sonnet 4.6 locally?
No. Claude Sonnet 4.6 is a closed, API-only model — the weights are not released, so it cannot be self-hosted.
Why does Claude Sonnet 4.6 charge more for output than input?
Output tokens are generated one at a time and cannot be batched the way a prompt can, so they cost the provider more to serve. Claude Sonnet 4.6 charges 5× more for output, which is why prompt-heavy workloads are far cheaper to run than generation-heavy ones.
Alle Claude-Modelle im direkten Preisvergleich: Claude-API-Preise.
Die Preise entsprechen den offiziell veröffentlichten Listenpreisen für die primäre API des jeweiligen Modells und werden regelmäßig aktualisiert, sobald Anbieter diese ändern. Volumen-, Batch- und Cached-Input-Rabatte sind nicht enthalten. Vergleichen Sie alle Modelle nebeneinander im Datenbank für KI-Modelle oder das LLM-Leaderboard.