Pass IndexThe State of AISign in

Claude Sonnet 4

Claude Sonnet 4 balances impressive performance for coding with the right speed and cost for high-volume use cases: Coding: Handle everyday development tasks with enhanced performance-power code reviews, bug fixes, API integrations, and feature development with immediate feedback loops.

text + image + file → text · made by Anthropic

$3$15per Mtok in / out
Anthropic · 12 sellers

Sold by 28 ways

SellerLaneRate
AWS Bedrock 0 ways$1.5per Mtok in$7.5per Mtok out
AWS Bedrock cloudnot on the seller's list todaybatch$1.5per Mtok in$7.5per Mtok outaws.amazon.com · read 2026-08-24
AWS Bedrock 0 ways$1.5per Mtok in$7.5per Mtok out
AWS Bedrock cloudnot on the seller's list todaybatch cross-region$1.5per Mtok in$7.5per Mtok outaws.amazon.com · read 2026-08-24
Anthropic apibatch$1.5per Mtok in$7.5per Mtok outplatform.claude.com · read 2026-08-24
Databricks cloudstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeraw.githubusercontent.com · read 2026-09-16
AWS Bedrock cloudstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeaws.amazon.com · read 2026-08-24
AWS Bedrock 0 ways$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache write
AWS Bedrock cloudnot on the seller's list todaycross-region$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeaws.amazon.com · read 2026-08-24
Anthropic apistandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeplatform.claude.com · read 2026-08-24
DigitalOcean cloudstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writemodels.dev · read 2026-09-16
Google Vertex AI cloudstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writemodels.dev · read 2026-09-16
Snowflake Cortex cloudstandard$3per Mtok in$15per Mtok out$0.3per Mtok cachedraw.githubusercontent.com · read 2026-09-16
AWS Bedrock cloudlong-context$6per Mtok in$22.5per Mtok out$0.6per Mtok cachedraw.githubusercontent.com · read 2026-09-16
Requesty 2 ways$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cached
Requesty aggregatornot on the seller's list todaybedrock eu-central-1$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todaybedrock eu-north-1$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todaybedrock eu-west-1$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todaybedrock eu-west-3$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todaybedrock us-east-1$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todaybedrock us-east-2$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todaybedrock us-west-2$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todayvertex europe-west1$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatornot on the seller's list todayvertex us-east5$2.7per Mtok in$13.5per Mtok out$0.27per Mtok cachedrouter.requesty.ai · read 2026-08-25
Requesty aggregatorbedrock$3per Mtok in$15per Mtok out$0.3per Mtok cachedrouter.requesty.ai · read 2026-08-28
Requesty aggregatorvertex$3per Mtok in$15per Mtok out$0.3per Mtok cachedrouter.requesty.ai · read 2026-08-28
ElectronHub aggregatorstandard$3per Mtok in$15per Mtok outapi.electronhub.ai · read 2026-09-16
GitHub Copilot 0 ways$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache write
GitHub Copilot aggregatornot on the seller's list todaystandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writemodels.dev · read 2026-09-04
Nous Research aggregatorstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeinference-api.nousresearch.com · read 2026-09-12
OpenRouter aggregatorstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeopenrouter.ai · read 2026-08-24
Vercel AI Gateway aggregatorstandard$3per Mtok in$15per Mtok out$0.3per Mtok cached$3.75per Mtok cache writeai-gateway.vercel.sh · read 2026-08-25
DeepInfra aggregatorstandard$3.3per Mtok in$16.5per Mtok outapi.deepinfra.com · read 2026-08-25

Measured 54 standings

PlaceBoardMetricScore
3rdof 22SWE-bench Multimodal% Resolved35.59
3rdof 16The agent company — Epoch AI% Score0.432
4thof 22SWE-bench Multimodal% Resolved (claude-sonnet 4)34.43
5thof 25SWE-Bench Pro (Public Dataset)Resolve Rate42.7
6thof 20Os world — Epoch AIScore43.9
6thof 180SWE-bench Verified (Bash Only toggle off — all agents)% Resolved76.8
8thof 25AgentBench Leaderboard (new)Success Rate (pass@1) AVG58.2
8thof 25AgentBench Leaderboard (new)Success Rate (pass@1) AVG (thinking)58.2
9thof 22Cybench — Epoch AIUnguided % Solved0.35
16thof 49Lech mazur writing — Epoch AIMean score8.14
20thof 28LiveCodeBench Leaderboard (default window 8/1/2024–5/1/2025, 454 problems)Pass@155.9
20thof 28LiveCodeBench Leaderboard (default window 8/1/2024–5/1/2025, 454 problems)Pass@1 (thinking)55.9
21stof 50Metr time horizons — Epoch AIaverage_score0.62
22ndof 69Aider polyglot coding leaderboardPercent correct61.3
22ndof 47SWE-bench Verified (default "Bash Only" view, agent = mini-SWE-agent)% Resolved64.93
25thof 72Aider polyglot — Epoch AIPercent correct61.3
25thof 38Gso — Epoch AIScore OPT@10.049
27thof 41Deepresearchbench — Epoch AIAverage score0.466
29thof 81Forecastbench — Epoch AIOverall score60.2
31stof 46Enigma eval — Epoch AIAccuracy3.12
32ndof 108Math level 5 — Epoch AImean_score84.366
32ndof 38Vpct — Epoch AICorrect0.34
36thof 42Fictionlivebench — Epoch AI120k token score0.364
37thof 51Humanity's Last Exam (Epoch AI replication)Accuracy7.76
52ndof 65Apex agents — Epoch AIPass@1 score0.093
58thof 105Vectara Hallucination LeaderboardHallucination Rate10.3
65thof 102Simplebench — Epoch AIScore (AVG@5)0.455
70thof 123Lmca — Epoch AIScore29
72ndof 72Frontiermath tier 4 — Epoch AImean_score0
74thof 101FrontierMath (Tiers 1-3)mean_score4.138
75thof 161Dtbench — Epoch AIAccuracy77.07
75thof 152LMArena · VisionScore (Elo)1206
77thof 112Ale bench — Epoch AIPerformance655.35
82ndof 147Design Arena · 3d (models)Elo1167
82ndof 170Weirdml — Epoch AIAccuracy46.11
85thof 109Design Arena · svg (models)Elo1103
93rdof 155Design Arena · gamedev (models)Elo1161
97thof 152Design Arena · dataviz (models)Elo1173
102ndof 150Design Arena · uicomponent (models)Elo1144
104thof 156Design Arena · codecategories (models)Elo1157
106thof 162Design Arena · website (models)Elo1158
114thof 168Scicode — Epoch AIScore40.046
121stof 184Artificial Analysis · Coding IndexIndex37.6
125thof 205ARC-AGI-1 (Public Eval)Score (%)56.75
126thof 177Critpt — Epoch AIAccuracy0.286
129thof 216ARC-AGI-2 (Public Eval)Score (%)4.44
130thof 290OTIS Mock AIME 2024-2025mean_score71.111
132ndof 232ARC-AGI-2 (Semi-Private)Score (%)5.93
132ndof 313GPQA Diamondmean_score78.283
134thof 227Arc agi 2 — Epoch AIScore5.93
148thof 402LMArena · TextScore (Elo)1401
150thof 230ARC-AGI-1 (Semi-Private)Score (%)40
161stof 244Arc agi — Epoch AIScore40
396thof 656Epoch capabilities index — Epoch AIECI Score141.72

About

Claude Sonnet 4 — a text model from Anthropic, sold by 12 companies from $3 in and $15 out per million tokens, placed 6th of 180 on SWE-bench Verified (Bash Only toggle off — all agents).

It takes text, images and files and returns text, with a context window of 1,000,000 tokens. It was published in May 2025, trained on material up to March 2025. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Twelve companies sell it. The cheapest is $3 in and $15 out per million tokens at Databricks. Beside the standard rate there are separately routed and batch lanes. It has been measured on 51 boards, and stands best at 6th of 180 on SWE-bench Verified (Bash Only toggle off — all agents).

Every current figure

Maker
Anthropic
Register
model
Takes
text + image + file
Returns
text
Context
1,000,000 tokens
Longest answer
200,000 tokens
Published
May 2025
Knowledge to
March 2025
Licence
not read
Sellers
12
Price
$3 in and $15 out per million tokens — Databricks
Boards
51
Best place
6th of 180 — SWE-bench Verified (Bash Only toggle off — all agents)

Known as 23 names

Claude Sonnet 4 (2025-05-14)Claude Sonnet 4 Thinking (2025-05-14)claude-sonnet-4-20250514 (32k thinking)Claude-Sonnet-4Claude-Sonnet-4 (Thinking)Claude 4 SonnetOpenHands-Versa (Claude-Sonnet 4)claude-4-SonnetEPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnetanthropic.claude-sonnet-4-20250514-v1:0anthropic/claude-4-sonnetanthropic/claude-sonnet-4bedrock/claude-sonnet-4bedrock/claude-sonnet-4@eu-central-1bedrock/claude-sonnet-4@eu-north-1bedrock/claude-sonnet-4@eu-west-1bedrock/claude-sonnet-4@eu-west-3bedrock/claude-sonnet-4@us-east-1bedrock/claude-sonnet-4@us-east-2bedrock/claude-sonnet-4@us-west-2vertex/claude-sonnet-4vertex/claude-sonnet-4@europe-west1vertex/claude-sonnet-4@us-east5