Verified November 2026

Home › Compare › Claude Sonnet 5.5 vs Claude Sonnet 4.6

Claude Sonnet 5.5 vs Claude Sonnet 4.6

A straight cost comparison of Claude Sonnet 5.5 (Anthropic) and Claude Sonnet 4.6 (Anthropic), per 1,000,000 tokens, verified November 2026.

$2Claude Sonnet 5.5 input / 1M
$3Claude Sonnet 4.6 input / 1M
33%Blended price gap
Claude Sonnet 5.5Cheaper blended

Rate-by-rate comparison

USD per 1M tokens · November 2026
MetricClaude Sonnet 5.5Claude Sonnet 4.6Winner
Input / 1M$2$3Claude Sonnet 5.5
Cached input / 1M$0.10$0.30Claude Sonnet 5.5
Cache write / 1M$2.50$3.75Claude Sonnet 5.5
Output / 1M$10$15Claude Sonnet 5.5
Context window1,000,0001,000,000Tie
Batch APIYesYesTie
Long-context tierNoneNone—

For per-token metrics, the lower number wins. For context window and batch support, the higher or broader option wins. A dash means the provider does not publish that rate: a missing cache-write fee means writes are not billed per token, while a missing cached-input rate means cache hits bill at the full input rate.

Blended cost: 1M input + 250K output

ModelBlended costDifference
Claude Sonnet 5.5$4.50Cheaper
Claude Sonnet 4.6$6.75+$2.25

Ten workloads, two bills

The same two models, priced across ten workloads that differ only in their input:output ratio. Nothing else changes — no retries, no system prompt, no middleware.

WorkloadClaude Sonnet 5.5Claude Sonnet 4.6Winner
Classification
100K reqs · 500 in / 10 out
$110$165Claude Sonnet 5.5
Data extraction
10K reqs · 2K in / 200 out
$60$90Claude Sonnet 5.5
Support chat
1K reqs · 4K in / 600 out
$14$21Claude Sonnet 5.5
Code review
2K reqs · 8K in / 1.5K out
$62$93Claude Sonnet 5.5
Summarisation
1K reqs · 20K in / 1K out
$50$75Claude Sonnet 5.5
Agent loop
5K reqs · 12K in / 2K out
$220$330Claude Sonnet 5.5
Translation
500 reqs · 15K in / 15K out
$90$135Claude Sonnet 5.5
Long-document QA
200 reqs · 120K in / 800 out
$49.60$74.40Claude Sonnet 5.5
Cached RAG
1K reqs · 50K in (45K cached) / 1K out
$24.61$43.66Claude Sonnet 5.5
Nightly batch
10K reqs · 10K in / 2K out
$200$300Claude Sonnet 5.5

Claude Sonnet 5.5 wins 10 of 10, Claude Sonnet 4.6 wins 0 of 10. The winner is set by the token mix, not by the headline rate.

Where the two bills cross over

Claude Sonnet 5.5 is cheaper on both sides of the mix: $2 vs $3 per million input tokens, and $10 vs $15 per million output. No input:output ratio reverses that, so on price alone the other model never becomes the better choice — it has to win on context, latency or output quality.

Cost at four scales

The same support-chat workload (4K input, 600 output per request) at four monthly volumes.

Requests / monthClaude Sonnet 5.5Claude Sonnet 4.6Difference
100$1.40$2.10Claude Sonnet 5.5 cheaper by $0.70
1,000$14$21Claude Sonnet 5.5 cheaper by $7
10,000$140$210Claude Sonnet 5.5 cheaper by $70
100,000$1,400$2,100Claude Sonnet 5.5 cheaper by $700

What caching does to each bill

Caching only helps the part of the prompt that repeats. This is the support-chat workload (1,000 requests, 4,000 input tokens each) with an increasing share of that prompt served from a warm cache, written once on the first request.

Cached share of the promptClaude Sonnet 5.5Claude Sonnet 4.6Winner
0%$14$21Claude Sonnet 5.5
50%$10.20$15.61Claude Sonnet 5.5
80%$7.93$12.37Claude Sonnet 5.5
95%$6.79$10.75Claude Sonnet 5.5

Both models publish a cached-read rate, so the discount is comparable: Claude Sonnet 5.5 reads cache at $0.10 per million, Claude Sonnet 4.6 at $0.30.

Which one should you pick?

Claude Sonnet 5.5 wins on classification, data extraction, support chat, code review, summarisation, agent loop, translation, long-document qa, cached rag, nightly batch, while Claude Sonnet 4.6 wins on none of the ten workloads above.

Beyond price, the two differ on input rate ($2 vs $3); output rate ($10 vs $15); cached-input rate ($0.10 vs $0.30); cache-write rate ($2.50 vs $3.75).

Batch mode is available on Claude Sonnet 5.5 and Claude Sonnet 4.6, which halves the nightly job. Both publish cached-input discounts, so a stable system prompt is worth caching on either.

FAQ

Is Claude Sonnet 5.5 cheaper than Claude Sonnet 4.6?

On a blended 1M-input-plus-250K-output workload, Claude Sonnet 5.5 is cheaper by 33% ($4.50 vs $6.75). The gap changes for cache-heavy prompts and for batch jobs, as the workload table shows.

Which handles longer prompts, Claude Sonnet 5.5 or Claude Sonnet 4.6?

Both models share the same 1,000,000-token context window. Neither model publishes a long-context surcharge.

Do both models support prompt caching?

Claude Sonnet 5.5 bills cached reads at $0.10 per 1M; Claude Sonnet 4.6 at $0.30.

Related comparisons

Run your own token volumes