Anthropic · Claude Sonnet · Frontier
Claude Sonnet 4 pricing and limits
The previous Sonnet generation at the same list price as Sonnet 4.5. Common in production systems pinned to a model snapshot; caching behaviour and cost drivers are the same.
| Line | Tokens | Rate / 1M | Cost |
|---|---|---|---|
| Input, uncached | 8,000 | $3.00 | $0.0240 |
| Input, cache read | 32,000 | $0.30 | $0.00960 |
| Output | 1,500 | $15.00 | $0.0225 |
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- Cached / 1M
- $0.30
- Cache write / 1M
- $3.75
- Blended 3:1
- $6.00
- Context
- 200K
What drives Claude Sonnet 4 cost
Prices match Sonnet 4.5, so migrating changes quality and behaviour, not the rate card. Cost differences between the two come from how many steps and output tokens each needs to finish a task, which is worth measuring per feature.
Cache writes at 1.25x input and reads at a tenth of input make prefix stability the main lever. Requests above 200K input tokens on the 1M context beta are billed at a higher rate.
Cost of a typical call
Take a agent step with 40,000 input tokens, 32,000 of them served from cache, and 1,500 output tokens. At Claude Sonnet 4 list prices ($3.00 input, $0.30 cached input, $15.00 output per 1M) one call costs $0.0561, against $0.1425 with no cache hits. At 80,000 calls a month that is $4,488, or $11,400 uncached.
Blended at a 3:1 input to output ratio, Claude Sonnet 4 costs $6.00 per 1M tokens. Routing the same call to the fallback in the policy below, Claude Haiku 4.5, costs $0.0187, which is what a tenant pays you in COGS once they cross the hard cap and PALM SOUQ downgrades instead of blocking.
Cost comparison
Claude Sonnet 4 against three alternatives
The same agent step priced on Claude Sonnet 4.5, GPT-4.1, Gemini 2.5 Pro. Move the sliders to match your traffic; models without a cache-read rate bill every input token in full.
Benchmark quality per task before switching. A cheaper model that needs more steps or longer outputs can cost more per finished task than its rate suggests, which is why PALM SOUQ reports cost per feature, not only per token.
Claude Sonnet 4 this guide
$0.0561 per call
$5,610 / mo
baseline
Claude Sonnet 4.5
$0.0561 per call
$5,610 / mo
+0% vs Claude Sonnet 4
GPT-4.1
$0.0440 per call
$4,400 / mo
-22% vs Claude Sonnet 4
Gemini 2.5 Pro
$0.0349 per call
$3,492 / mo
-38% vs Claude Sonnet 4
List prices from the PALM SOUQ LLM Price Index, computed per call as uncached input + cache reads + output. Reasoning tokens count as output on reasoning models.
Rate limits by usage tier
Per-model RPM, input TPM and output TPM by usage tier. Limits are tracked per model, so running Sonnet 4 and 4.5 side by side does not pool their capacity.
Exact limits vary by account tier and change over time, so we do not reprint them here. Check the limits page for your own account in the Anthropic documentation at docs.anthropic.com. PALM SOUQ enforces your own per-tenant limits in front of the provider's, so a single customer cannot consume the shared allowance that every other tenant depends on.
Circuit-breaker best practices
Set the max step cap from observed data: PALM SOUQ records steps per session per tenant, and the 99th percentile plus a margin is a better cap than a guess.
Use retry storm detection for 429 and 529 responses. Clients that retry immediately on overload multiply cost and make the overload worse.
When a breaker trips, the gateway answers in under 50 ms with HTTP 429 and X-PalmSouq-Reason: circuit_breaker_triggered, before the request reaches Anthropic. Your agent framework sees a normal rate-limit response and stops, and the audit log records which heuristic fired and for which tenant.
# palmsouq.policy.yaml
policy: claude-sonnet-4-pro
match:
model: claude-sonnet-4
tier: pro # from X-PalmSouq-Tier
budgets:
monthly_usd: 400 # per tenant
soft_cap_pct: 80 # webhook to your app
on_hard_cap: fallback # or: block (HTTP 429)
fallback:
- claude-haiku-4.5
breakers:
tool_thrashing: { identical_calls: 3, window_steps: 8 }
token_velocity: { max_tokens_per_s: 50000, window_s: 10 }
semantic_echo: { repeats: 3 }
max_steps: 25
cache:
meter_reads: true
meter_writes: true
alert_read_share_below: 0.6- budgetsPer-tenant monthly dollar cap with an 80% soft-cap webhook, keyed by X-PalmSouq-Tenant.
- fallbackAt the hard cap, requests reroute to Claude Haiku 4.5 ($2.00 blended per 1M) instead of failing.
- breakersAll four loop heuristics on, evaluated in the request path with under 5 ms added latency.
Where it fits
Stable production workloads that were evaluated on this snapshot. Plan a migration test to Sonnet 4.5 using per-feature cost and step counts as the comparison.
Provider documentation: docs.anthropic.com
Claude Sonnet 4 pricing questions
How much does Claude Sonnet 4 cost per million tokens?
At list price Claude Sonnet 4 costs $3.00 per 1M input tokens and $15.00 per 1M output tokens, with cached input at $0.30 per 1M and five-minute cache writes at $3.75 per 1M. Prices were last reviewed in the PALM SOUQ LLM Price Index and are for the standard tier and shortest context band.
Does Claude Sonnet 4 support prompt caching?
Yes. Cache reads cost $0.30 per 1M tokens and cache writes cost $3.75 per 1M for the five-minute TTL. Caching saves money only when a prefix is read more often than it is rewritten, which is why PALM SOUQ meters reads and writes separately per tenant.
What are the Claude Sonnet 4 rate limits?
Per-model RPM, input TPM and output TPM by usage tier. Limits are tracked per model, so running Sonnet 4 and 4.5 side by side does not pool their capacity. For your account's exact numbers, see the Anthropic documentation at docs.anthropic.com.
How do I stop runaway Claude Sonnet 4 agent loops?
Set the max step cap from observed data: PALM SOUQ records steps per session per tenant, and the 99th percentile plus a margin is a better cap than a guess. The sample policy on this page enables all four PALM SOUQ heuristics with a max step cap of 25 and falls back to Claude Haiku 4.5 at the hard cap.
Is there a batch discount for Claude Sonnet 4?
50% off input and output through the Message Batches API. Batch traffic suits evaluations, backfills and nightly enrichment jobs that do not need a real-time answer.
Govern Claude Sonnet 4 spend per customer.
Point your Anthropic client at gateway.palmsouqltd.com, pass X-PalmSouq-Tenant, and every Claude Sonnet 4 token is attributed, capped and protected by circuit breakers.
14-day free production trial · 100k requests included · no card required