Claude Sonnet 5

anthropic/claude-sonnet-5-20260630

Claude Sonnet 5 is an Anthropic model that accepts text, image, and file inputs and produces text output. It supports tool usage and reasoning, with a

tool use
structured outputs
reasoning

At a glance

provideranthropic

releasedJun 29, 2026

context1M

input / M$2

Specifications

What the API gives you

Context window: 1M tokens
Max output: 128K tokens
Input price: $2 / 1M tokens
Output price: $10 / 1M tokens
Cache read: $0.20 / 1M tokens
Modalities: text, image, file → text
Knowledge cutoff: —
Released: Jun 29, 2026

Measured evidence

Public benchmark record

Design Arena Coding: 1304 Elo2026-07-28
Design Arena Coding: 1312 Elo2026-07-27
Design Arena Coding: 1303 Elo2026-07-30
Design Arena Coding: 1304 Elo2026-07-29
Design Arena Coding: 1302 Elo2026-07-31
Design Arena Coding: 1301 Elo2026-08-02
Design Arena Coding: 1300 Elo2026-08-05
Design Arena Coding: 1301 Elo2026-08-01
Design Arena Coding: 1300 Elo2026-08-09
Design Arena Coding: 1301 Elo2026-08-03
Design Arena Coding: 1301 Elo2026-08-04
Design Arena Coding: 1299 Elo2026-08-14
Design Arena Coding: 1301 Elo2026-08-06
Design Arena Coding: 1300 Elo2026-08-10
Design Arena Coding: 1301 Elo2026-08-07
Design Arena Coding: 1295 Elo2026-08-18
Design Arena Coding: 1301 Elo2026-08-08
Design Arena Coding: 1300 Elo2026-08-21
Design Arena Coding: 1295 Elo2026-08-15
Design Arena Coding: 1301 Elo2026-08-11
Design Arena Coding: 1300 Elo2026-08-12
Design Arena Coding: 1299 Elo2026-08-13
Design Arena Coding: 1294 Elo2026-08-19
Design Arena Coding: 1300 Elo2026-08-27
Design Arena Coding: 1294 Elo2026-08-20
Design Arena Coding: 1296 Elo2026-08-16
Design Arena Coding: 1292 Elo2026-09-05
Design Arena Coding: 1295 Elo2026-08-17
Design Arena Coding: 1301 Elo2026-08-22
Design Arena Coding: 1301 Elo2026-08-23
Design Arena Coding: 1300 Elo2026-08-28
Design Arena Coding: 1301 Elo2026-08-24
Design Arena Coding: 1301 Elo2026-08-25
Design Arena Coding: 1301 Elo2026-08-26
Design Arena Coding: 1293 Elo2026-09-06
Design Arena Coding: 1300 Elo2026-08-29
Design Arena Coding: 1299 Elo2026-08-30
Design Arena Coding: 1299 Elo2026-08-31
Design Arena Coding: 1299 Elo2026-09-01
Design Arena Coding: 1298 Elo2026-09-02
Design Arena Coding: 1296 Elo2026-09-03
Design Arena Coding: 1295 Elo2026-09-04
Design Arena Full-stack Agent: 1275 Elo2026-07-28
Design Arena Full-stack Agent: 1275 Elo2026-07-27
Design Arena Full-stack Agent: 1268 Elo2026-07-30
Design Arena Full-stack Agent: 1270 Elo2026-07-29
Design Arena Full-stack Agent: 1269 Elo2026-07-31
Design Arena Full-stack Agent: 1271 Elo2026-08-02
Design Arena Full-stack Agent: 1272 Elo2026-08-05
Design Arena Full-stack Agent: 1269 Elo2026-08-01
Design Arena Full-stack Agent: 1264 Elo2026-08-09
Design Arena Full-stack Agent: 1273 Elo2026-08-03
Design Arena Full-stack Agent: 1272 Elo2026-08-04
Design Arena Full-stack Agent: 1271 Elo2026-08-14
Design Arena Full-stack Agent: 1272 Elo2026-08-06
Design Arena Full-stack Agent: 1260 Elo2026-08-10
Design Arena Full-stack Agent: 1272 Elo2026-08-07
Design Arena Full-stack Agent: 1274 Elo2026-08-18
Design Arena Full-stack Agent: 1274 Elo2026-08-08
Design Arena Full-stack Agent: 1276 Elo2026-08-15
Design Arena Full-stack Agent: 1265 Elo2026-08-11
Design Arena Full-stack Agent: 1267 Elo2026-08-12
Design Arena Full-stack Agent: 1270 Elo2026-08-13
Design Arena Full-stack Agent: 1274 Elo2026-08-19
Design Arena Full-stack Agent: 1269 Elo2026-08-27
Design Arena Full-stack Agent: 1274 Elo2026-08-20
Design Arena Full-stack Agent: 1280 Elo2026-08-16
Design Arena Full-stack Agent: 1260 Elo2026-09-05
Design Arena Full-stack Agent: 1279 Elo2026-08-17
Design Arena Full-stack Agent: 1282 Elo2026-08-21
Design Arena Full-stack Agent: 1282 Elo2026-08-22
Design Arena Full-stack Agent: 1281 Elo2026-08-23
Design Arena Full-stack Agent: 1268 Elo2026-08-28
Design Arena Full-stack Agent: 1273 Elo2026-08-24
Design Arena Full-stack Agent: 1271 Elo2026-08-25
Design Arena Full-stack Agent: 1271 Elo2026-08-26
Design Arena Full-stack Agent: 1258 Elo2026-09-06
Design Arena Full-stack Agent: 1268 Elo2026-08-29
Design Arena Full-stack Agent: 1270 Elo2026-08-30
Design Arena Full-stack Agent: 1268 Elo2026-08-31
Design Arena Full-stack Agent: 1270 Elo2026-09-01
Design Arena Full-stack Agent: 1270 Elo2026-09-02
Design Arena Full-stack Agent: 1271 Elo2026-09-03
Design Arena Full-stack Agent: 1271 Elo2026-09-04
Design Arena UI: 1308 Elo2026-07-28
Design Arena UI: 1317 Elo2026-07-27
Design Arena UI: 1308 Elo2026-07-30
Design Arena UI: 1308 Elo2026-07-29
Design Arena UI: 1308 Elo2026-07-31
Design Arena UI: 1309 Elo2026-08-02
Design Arena UI: 1309 Elo2026-08-01
Design Arena UI: 1311 Elo2026-08-05
Design Arena UI: 1310 Elo2026-08-09
Design Arena UI: 1310 Elo2026-08-03
Design Arena UI: 1312 Elo2026-08-04
Design Arena UI: 1309 Elo2026-08-14
Design Arena UI: 1313 Elo2026-08-06
Design Arena UI: 1313 Elo2026-08-07
Design Arena UI: 1310 Elo2026-08-10
Design Arena UI: 1311 Elo2026-08-08
Design Arena UI: 1307 Elo2026-08-18
Design Arena UI: 1309 Elo2026-08-11
Design Arena UI: 1308 Elo2026-08-15
Design Arena UI: 1311 Elo2026-08-12
Design Arena UI: 1311 Elo2026-08-13
Design Arena UI: 1307 Elo2026-08-19
Design Arena UI: 1305 Elo2026-08-27
Design Arena UI: 1309 Elo2026-08-16
Design Arena UI: 1307 Elo2026-08-20
Design Arena UI: 1308 Elo2026-08-17
Design Arena UI: 1302 Elo2026-09-05
Design Arena UI: 1304 Elo2026-08-21
Design Arena UI: 1306 Elo2026-08-22
Design Arena UI: 1306 Elo2026-08-23
Design Arena UI: 1305 Elo2026-08-28
Design Arena UI: 1306 Elo2026-08-24
Design Arena UI: 1306 Elo2026-08-25
Design Arena UI: 1306 Elo2026-08-26
Design Arena UI: 1303 Elo2026-09-06
Design Arena UI: 1305 Elo2026-08-29
Design Arena UI: 1305 Elo2026-08-30
Design Arena UI: 1304 Elo2026-08-31
Design Arena UI: 1305 Elo2026-09-01
Design Arena UI: 1305 Elo2026-09-02
Design Arena UI: 1303 Elo2026-09-03
Design Arena UI: 1302 Elo2026-09-04
Design Arena Web-app Agent: 1301 Elo2026-07-28
Design Arena Web-app Agent: 1301 Elo2026-07-27
Design Arena Web-app Agent: 1304 Elo2026-07-30
Design Arena Web-app Agent: 1303 Elo2026-07-29
Design Arena Web-app Agent: 1304 Elo2026-07-31
Design Arena Web-app Agent: 1303 Elo2026-08-02
Design Arena Web-app Agent: 1303 Elo2026-08-01
Design Arena Web-app Agent: 1300 Elo2026-08-05
Design Arena Web-app Agent: 1295 Elo2026-08-09
Design Arena Web-app Agent: 1302 Elo2026-08-03
Design Arena Web-app Agent: 1302 Elo2026-08-04
Design Arena Web-app Agent: 1291 Elo2026-08-14
Design Arena Web-app Agent: 1300 Elo2026-08-06
Design Arena Web-app Agent: 1293 Elo2026-08-10
Design Arena Web-app Agent: 1300 Elo2026-08-07
Design Arena Web-app Agent: 1286 Elo2026-08-18
Design Arena Web-app Agent: 1301 Elo2026-08-08
Design Arena Web-app Agent: 1294 Elo2026-08-11
Design Arena Web-app Agent: 1288 Elo2026-08-15
Design Arena Web-app Agent: 1294 Elo2026-08-12
Design Arena Web-app Agent: 1292 Elo2026-08-13
Design Arena Web-app Agent: 1286 Elo2026-08-19
Design Arena Web-app Agent: 1275 Elo2026-08-27
Design Arena Web-app Agent: 1286 Elo2026-08-20
Design Arena Web-app Agent: 1291 Elo2026-08-16
Design Arena Web-app Agent: 1277 Elo2026-09-05
Design Arena Web-app Agent: 1291 Elo2026-08-17
Design Arena Web-app Agent: 1285 Elo2026-08-21
Design Arena Web-app Agent: 1282 Elo2026-08-22
Design Arena Web-app Agent: 1282 Elo2026-08-23
Design Arena Web-app Agent: 1277 Elo2026-08-28
Design Arena Web-app Agent: 1276 Elo2026-08-24
Design Arena Web-app Agent: 1277 Elo2026-08-25
Design Arena Web-app Agent: 1276 Elo2026-08-26
Design Arena Web-app Agent: 1277 Elo2026-09-06
Design Arena Web-app Agent: 1278 Elo2026-08-29
Design Arena Web-app Agent: 1278 Elo2026-08-30
Design Arena Web-app Agent: 1278 Elo2026-08-31
Design Arena Web-app Agent: 1278 Elo2026-09-01
Design Arena Web-app Agent: 1277 Elo2026-09-02
Design Arena Web-app Agent: 1277 Elo2026-09-03
Design Arena Web-app Agent: 1277 Elo2026-09-04
Design Arena Website: 1302 Elo2026-07-28
Design Arena Website: 1300 Elo2026-07-30
Design Arena Website: 1314 Elo2026-07-27
Design Arena Website: 1301 Elo2026-07-29
Design Arena Website: 1299 Elo2026-07-31
Design Arena Website: 1298 Elo2026-08-02
Design Arena Website: 1298 Elo2026-08-01
Design Arena Website: 1296 Elo2026-08-05
Design Arena Website: 1297 Elo2026-08-03
Design Arena Website: 1297 Elo2026-08-09
Design Arena Website: 1297 Elo2026-08-04
Design Arena Website: 1295 Elo2026-08-14
Design Arena Website: 1296 Elo2026-08-06
Design Arena Website: 1296 Elo2026-08-07
Design Arena Website: 1297 Elo2026-08-10
Design Arena Website: 1297 Elo2026-08-08
Design Arena Website: 1286 Elo2026-08-18
Design Arena Website: 1298 Elo2026-08-11
Design Arena Website: 1286 Elo2026-08-15
Design Arena Website: 1296 Elo2026-08-12
Design Arena Website: 1295 Elo2026-08-13
Design Arena Website: 1286 Elo2026-08-19
Design Arena Website: 1286 Elo2026-08-16
Design Arena Website: 1286 Elo2026-08-20
Design Arena Website: 1297 Elo2026-08-27
Design Arena Website: 1286 Elo2026-08-17
Design Arena Website: 1292 Elo2026-09-05
Design Arena Website: 1296 Elo2026-08-21
Design Arena Website: 1297 Elo2026-08-22
Design Arena Website: 1297 Elo2026-08-23
Design Arena Website: 1297 Elo2026-08-24
Design Arena Website: 1297 Elo2026-08-28
Design Arena Website: 1297 Elo2026-08-25
Design Arena Website: 1297 Elo2026-08-26
Design Arena Website: 1290 Elo2026-09-06
Design Arena Website: 1296 Elo2026-08-29
Design Arena Website: 1295 Elo2026-08-30
Design Arena Website: 1295 Elo2026-08-31
Design Arena Website: 1295 Elo2026-09-01
Design Arena Website: 1294 Elo2026-09-02
Design Arena Website: 1293 Elo2026-09-03
Design Arena Website: 1292 Elo2026-09-04
Frontier-Bench^‡: 14.6% ± 1.52026-07-12Claude Code, max effort
Frontier-Bench^‡: 12.4%2026-06-30Claude Code, max effort
LiveBench Agentic Coding: 59.4%2026-06-25
LiveBench Coding: 80.7%2026-06-25
LiveBench Data Analysis: 71.7%2026-06-25
LiveBench Instruction Following: 63.9%2026-06-25
LiveBench Language: 75.0%2026-06-25
LiveBench Math: 92.9%2026-06-25
LiveBench Reasoning: 88.7%2026-06-25
LMArena Agent: 82026-07-13
LMArena Agent: 9.12026-07-20
LMArena Agent: 8.62026-07-27
LMArena Agent: 8.72026-07-21
LMArena Agent: 8.62026-07-28
LMArena Agent: 7.52026-08-03
LMArena Agent: 7.32026-08-12
LMArena Agent: 7.52026-08-06
LMArena Agent: 7.42026-08-11
LMArena Agent: 7.12026-08-13
LMArena Agent: 82026-08-24
LMArena Agent: 7.52026-08-31
LMArena Agent: 7.42026-09-05
LMArena Agent: 7.72026-08-27
LMArena Creative Writing: 1409 Elo2026-07-21
LMArena Creative Writing: 1407 Elo2026-07-27
LMArena Creative Writing: 1407 Elo2026-07-30
LMArena Creative Writing: 1410 Elo2026-08-02
LMArena Creative Writing: 1410 Elo2026-08-03
LMArena Creative Writing: 1411 Elo2026-08-06
LMArena Creative Writing: 1415 Elo2026-08-12
LMArena Creative Writing: 1417 Elo2026-08-21
LMArena Creative Writing: 1418 Elo2026-09-02
LMArena Creative Writing: 1418 Elo2026-08-27
LMArena Creative Writing: 1418 Elo2026-09-01
LMArena Instruction Following: 1445 Elo2026-07-21
LMArena Instruction Following: 1445 Elo2026-07-27
LMArena Instruction Following: 1445 Elo2026-07-30
LMArena Instruction Following: 1448 Elo2026-08-02
LMArena Instruction Following: 1448 Elo2026-08-03
LMArena Instruction Following: 1450 Elo2026-08-06
LMArena Instruction Following: 1453 Elo2026-08-12
LMArena Instruction Following: 1453 Elo2026-08-21
LMArena Instruction Following: 1451 Elo2026-09-02
LMArena Instruction Following: 1451 Elo2026-08-27
LMArena Instruction Following: 1451 Elo2026-09-01
LMArena Long Query: 1462 Elo2026-07-21
LMArena Long Query: 1460 Elo2026-07-27
LMArena Long Query: 1460 Elo2026-07-30
LMArena Long Query: 1463 Elo2026-08-02
LMArena Long Query: 1463 Elo2026-08-03
LMArena Long Query: 1464 Elo2026-08-06
LMArena Long Query: 1464 Elo2026-08-12
LMArena Long Query: 1464 Elo2026-08-21
LMArena Long Query: 1464 Elo2026-09-02
LMArena Long Query: 1464 Elo2026-08-27
LMArena Long Query: 1464 Elo2026-09-01
LMArena Math: 1461 Elo2026-07-21
LMArena Math: 1463 Elo2026-07-27
LMArena Math: 1463 Elo2026-07-30
LMArena Math: 1464 Elo2026-08-02
LMArena Math: 1464 Elo2026-08-03
LMArena Math: 1465 Elo2026-08-06
LMArena Math: 1462 Elo2026-08-12
LMArena Math: 1463 Elo2026-08-21
LMArena Math: 1468 Elo2026-09-01
LMArena Math: 1465 Elo2026-08-27
LMArena Text: 1442 Elo2026-07-16
LMArena Text: 1462 Elo2026-08-02
LMArena Text: 1461 Elo2026-07-21
LMArena Text: 1460 Elo2026-07-27
LMArena Text: 1460 Elo2026-07-30
LMArena Text: 1463 Elo2026-08-06
LMArena Text: 1462 Elo2026-08-03
LMArena Text: 1463 Elo2026-08-10
LMArena Text: 1461 Elo2026-08-21
LMArena Text: 1462 Elo2026-08-12
LMArena Text: 1462 Elo2026-09-01
LMArena Text: 1462 Elo2026-09-02
LMArena Text: 1461 Elo2026-08-27
LMArena Vision: 1275 Elo2026-07-12
LMArena Vision: 1272 Elo2026-07-21
LMArena Vision: 1271 Elo2026-07-26
LMArena Vision: 1271 Elo2026-07-30
LMArena Vision: 1275 Elo2026-08-02
LMArena Vision: 1273 Elo2026-08-06
LMArena Vision: 1271 Elo2026-08-21
LMArena Vision: 1267 Elo2026-08-27
LMArena WebDev: 1542 Elo2026-07-16
LMArena WebDev: 1544 Elo2026-07-21
LMArena WebDev: 1546 Elo2026-07-20
LMArena WebDev: 1545 Elo2026-07-27
LMArena WebDev: 1541 Elo2026-07-24
LMArena WebDev: 1544 Elo2026-07-28
LMArena WebDev: 1544 Elo2026-07-30
LMArena WebDev: 1542 Elo2026-08-01
LMArena WebDev: 1543 Elo2026-07-31
LMArena WebDev: 1541 Elo2026-08-06
LMArena WebDev: 1542 Elo2026-08-02
LMArena WebDev: 1541 Elo2026-08-12
LMArena WebDev: 1542 Elo2026-08-10
LMArena WebDev: 1540 Elo2026-08-14
LMArena WebDev: 1537 Elo2026-09-01
LMArena WebDev: 1537 Elo2026-09-05
LMArena WebDev: 1539 Elo2026-08-21
LMArena WebDev: 1537 Elo2026-08-30
SWE-rebench^‡: 56.8% ± 0.92026-07-01tools
Terminal-Bench 2.1^‡: 74.6% ± 1.62026-07-09high effort
Terminal-Bench 2.1^‡: 12.4%2026-06-30max effort

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference. ‡ Submitted scaffold plus model result. Compare the harness and effort settings, not only the model name.

Task evidence

How broad is the support?

Each status is exact to this model version and the selected task. Official release claims remain provisional.

Best LLM for Competition MathEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Business WritingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Agentic CodingEstablished evidence
100% of intended weighted task evidence across 6 independent sources. 8 of 8 configured evaluation feeds currently have results.
Best LLM for Logic & Problem SolvingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Marketing CopyEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Chat & RoleplayEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Creative WritingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Frontend & UIEstablished evidence
100% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for WritingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Roleplay & CharacterEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Math & ReasoningEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for a General AssistantEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Autonomous AgentsEstablished evidence
95% of intended weighted task evidence across 5 independent sources. 7 of 7 configured evaluation feeds currently have results.
Best LLM for CodingEstablished evidence
84% of intended weighted task evidence across 5 independent sources. 8 of 8 configured evaluation feeds currently have results.
Best LLM for Code CompletionEstablished evidence
81% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
Best LLM for Debugging & Code ReviewEstablished evidence
70% of intended weighted task evidence across 3 independent sources. 6 of 6 configured evaluation feeds currently have results.
Best Open-Weight LLM for Self-HostingEstablished evidence
67% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
Best LLM for Image UnderstandingEstablished evidence
65% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
Best Local LLM for CodingEstablished evidence
64% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Fast SummariesEstablished evidence
64% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
Best Value LLMEstablished evidence
64% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
Best LLM for Hermes AgentEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for OpenHandsEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for GooseEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for OpenClawEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
Best LLM for Document SummarizationEstablished evidence
60% of intended weighted task evidence across 2 independent sources. 4 of 6 configured evaluation feeds currently have results.
Best LLM for SummarizationEstablished evidence
60% of intended weighted task evidence across 2 independent sources. 4 of 6 configured evaluation feeds currently have results.
Best LLM for AI AgentsProvisional evidence
60% of intended weighted task evidence across 4 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or OSWorld Verified. 7 of 7 configured evaluation feeds currently have results. Official release evidence is also tracked provisionally.
Best LLM for Literary TranslationProvisional evidence
55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from WMT translation. 4 of 5 configured evaluation feeds currently have results.
Best LLM for Everyday TranslationProvisional evidence
55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from WMT translation. 4 of 5 configured evaluation feeds currently have results.
Best LLM for TranslationProvisional evidence
55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from WMT translation. 4 of 5 configured evaluation feeds currently have results.
Best Cheap LLM for High VolumeProvisional evidence
50% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Route reliability. 3 of 4 configured evaluation feeds currently have results. Official release evidence is also tracked provisionally.
Cheapest Capable LLMProvisional evidence
50% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Route reliability. 3 of 4 configured evaluation feeds currently have results.
Best LLM for Massive ContextProvisional evidence
50% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LongBench v2. 3 of 4 configured evaluation feeds currently have results.
Best LLM for OCR & DocumentsProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
Best LLM for Vision & DocumentsProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
Best LLM for Charts & DiagramsProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
Best LLM for Document ParsingProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
Best LLM for JSON & Schema OutputProvisional evidence
39% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.
Best LLM for Structured Data ExtractionProvisional evidence
39% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.
Best LLM for RAG & Long ContextProvisional evidence
35% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Search. 4 of 6 configured evaluation feeds currently have results. Official release evidence is also tracked provisionally.
Best LLM for High-Volume ExtractionProvisional evidence
31% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Route reliability or Structured-output eval. 2 of 4 configured evaluation feeds currently have results.
Best LLM for Tool & Function CallingProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Search. 4 of 5 configured evaluation feeds currently have results. Official release evidence is also tracked provisionally.
Best LLM for Computer & Browser UseProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or OSWorld Verified. 4 of 4 configured evaluation feeds currently have results. Official release evidence is also tracked provisionally.
Best LLM for Grounded RAGProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from CRAG or FACTS Grounding. 2 of 5 configured evaluation feeds currently have results.

Official release evidence · provisional

What the launch material actually says

Verbatim claims from the model provider. These help on launch day but do not count as independent confirmation.

“It's a substantial improvement over its predecessor, Sonnet 4.6, on important aspects of agentic performance like reasoning, tool use, coding, and knowledge work”
Anthropic · Introducing Claude Sonnet 5 \ Anthropic
“Claude Sonnet 5 gets more done with less. Same output quality, fewer steps to get there.”
Anthropic · Introducing Claude Sonnet 5 \ Anthropic
“Sonnet 5 narrows the gap: its performance is close to that of Opus 4.8, but at lower prices.”
Anthropic · Introducing Claude Sonnet 5 \ Anthropic
“Claude Sonnet 5 gives our agents a strong execution layer for multi-step software engineering work. It handles sustained coding, tool use, and debugging well across messy technical contexts”
Anthropic · Introducing Claude Sonnet 5 \ Anthropic

Recommendation coverage

Where this model appears

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.