Claude Opus 5

anthropic/claude-opus-5-20260723

Claude Opus 5 is a multimodal AI model from Anthropic that accepts text, image, and file inputs and generates text outputs. It has a context length of 1,000,000 tokens, supports tool use and reasoning, and is a proprietary model released in July 2026.

tool use
structured outputs
reasoning

At a glance

provideranthropic

releasedJul 23, 2026

context1M

input / M$5

Specifications

What the API gives you

Context window: 1M tokens
Max output: 128K tokens
Input price: $5 / 1M tokens
Output price: $25 / 1M tokens
Cache read: $0.50 / 1M tokens
Modalities: text, image, file → text
Knowledge cutoff: —
Released: Jul 23, 2026

Measured evidence

Public benchmark record

Design Arena Coding: 1358 Elo2026-07-28
Design Arena Coding: 1360 Elo2026-07-27
Design Arena Coding: 1358 Elo2026-07-30
Design Arena Coding: 1358 Elo2026-07-29
Design Arena Coding: 1357 Elo2026-07-31
Design Arena Coding: 1355 Elo2026-08-02
Design Arena Coding: 1361 Elo2026-08-05
Design Arena Coding: 1355 Elo2026-08-01
Design Arena Coding: 1358 Elo2026-08-09
Design Arena Coding: 1356 Elo2026-08-03
Design Arena Coding: 1362 Elo2026-08-04
Design Arena Coding: 1355 Elo2026-08-14
Design Arena Coding: 1362 Elo2026-08-06
Design Arena Coding: 1358 Elo2026-08-10
Design Arena Coding: 1361 Elo2026-08-07
Design Arena Coding: 1347 Elo2026-08-18
Design Arena Coding: 1359 Elo2026-08-08
Design Arena Coding: 1350 Elo2026-08-21
Design Arena Coding: 1348 Elo2026-08-15
Design Arena Coding: 1358 Elo2026-08-11
Design Arena Coding: 1357 Elo2026-08-12
Design Arena Coding: 1344 Elo2026-08-19
Design Arena Coding: 1354 Elo2026-08-13
Design Arena Coding: 1349 Elo2026-08-27
Design Arena Coding: 1341 Elo2026-09-05
Design Arena Coding: 1344 Elo2026-08-20
Design Arena Coding: 1347 Elo2026-08-16
Design Arena Coding: 1347 Elo2026-08-17
Design Arena Coding: 1350 Elo2026-08-22
Design Arena Coding: 1350 Elo2026-08-23
Design Arena Coding: 1348 Elo2026-08-28
Design Arena Coding: 1350 Elo2026-08-24
Design Arena Coding: 1349 Elo2026-08-25
Design Arena Coding: 1349 Elo2026-08-26
Design Arena Coding: 1340 Elo2026-09-06
Design Arena Coding: 1347 Elo2026-08-29
Design Arena Coding: 1346 Elo2026-08-30
Design Arena Coding: 1340 Elo2026-09-07
Design Arena Coding: 1346 Elo2026-08-31
Design Arena Coding: 1346 Elo2026-09-01
Design Arena Coding: 1345 Elo2026-09-02
Design Arena Coding: 1341 Elo2026-09-08
Design Arena Coding: 1342 Elo2026-09-03
Design Arena Coding: 1343 Elo2026-09-04
Design Arena Full-stack Agent: 1347 Elo2026-08-18
Design Arena Full-stack Agent: 1347 Elo2026-08-19
Design Arena Full-stack Agent: 1347 Elo2026-08-27
Design Arena Full-stack Agent: 1347 Elo2026-08-20
Design Arena Full-stack Agent: 1352 Elo2026-08-16
Design Arena Full-stack Agent: 1352 Elo2026-09-05
Design Arena Full-stack Agent: 1352 Elo2026-08-17
Design Arena Full-stack Agent: 1354 Elo2026-08-21
Design Arena Full-stack Agent: 1354 Elo2026-08-22
Design Arena Full-stack Agent: 1354 Elo2026-08-23
Design Arena Full-stack Agent: 1349 Elo2026-08-28
Design Arena Full-stack Agent: 1345 Elo2026-08-24
Design Arena Full-stack Agent: 1347 Elo2026-08-25
Design Arena Full-stack Agent: 1348 Elo2026-08-26
Design Arena Full-stack Agent: 1350 Elo2026-09-06
Design Arena Full-stack Agent: 1348 Elo2026-08-29
Design Arena Full-stack Agent: 1357 Elo2026-08-30
Design Arena Full-stack Agent: 1347 Elo2026-09-07
Design Arena Full-stack Agent: 1358 Elo2026-08-31
Design Arena Full-stack Agent: 1360 Elo2026-09-01
Design Arena Full-stack Agent: 1361 Elo2026-09-02
Design Arena Full-stack Agent: 1363 Elo2026-09-03
Design Arena Full-stack Agent: 1352 Elo2026-09-08
Design Arena Full-stack Agent: 1363 Elo2026-09-04
Design Arena UI: 1402 Elo2026-07-28
Design Arena UI: 1396 Elo2026-07-27
Design Arena UI: 1405 Elo2026-07-30
Design Arena UI: 1405 Elo2026-07-29
Design Arena UI: 1404 Elo2026-07-31
Design Arena UI: 1403 Elo2026-08-02
Design Arena UI: 1403 Elo2026-08-01
Design Arena UI: 1385 Elo2026-08-05
Design Arena UI: 1383 Elo2026-08-09
Design Arena UI: 1400 Elo2026-08-03
Design Arena UI: 1387 Elo2026-08-04
Design Arena UI: 1378 Elo2026-08-14
Design Arena UI: 1386 Elo2026-08-06
Design Arena UI: 1382 Elo2026-08-10
Design Arena UI: 1384 Elo2026-08-07
Design Arena UI: 1384 Elo2026-08-08
Design Arena UI: 1368 Elo2026-08-18
Design Arena UI: 1383 Elo2026-08-11
Design Arena UI: 1370 Elo2026-08-15
Design Arena UI: 1381 Elo2026-08-12
Design Arena UI: 1381 Elo2026-08-13
Design Arena UI: 1368 Elo2026-08-19
Design Arena UI: 1373 Elo2026-08-27
Design Arena UI: 1368 Elo2026-08-20
Design Arena UI: 1370 Elo2026-08-16
Design Arena UI: 1369 Elo2026-08-17
Design Arena UI: 1368 Elo2026-09-05
Design Arena UI: 1372 Elo2026-08-21
Design Arena UI: 1373 Elo2026-08-22
Design Arena UI: 1373 Elo2026-08-23
Design Arena UI: 1373 Elo2026-08-28
Design Arena UI: 1371 Elo2026-08-24
Design Arena UI: 1372 Elo2026-08-25
Design Arena UI: 1373 Elo2026-08-26
Design Arena UI: 1368 Elo2026-09-06
Design Arena UI: 1373 Elo2026-08-29
Design Arena UI: 1372 Elo2026-08-30
Design Arena UI: 1368 Elo2026-09-07
Design Arena UI: 1372 Elo2026-08-31
Design Arena UI: 1371 Elo2026-09-01
Design Arena UI: 1372 Elo2026-09-02
Design Arena UI: 1370 Elo2026-09-03
Design Arena UI: 1366 Elo2026-09-08
Design Arena UI: 1367 Elo2026-09-04
Design Arena Web-app Agent: 1293 Elo2026-07-28
Design Arena Web-app Agent: 1292 Elo2026-07-27
Design Arena Web-app Agent: 1294 Elo2026-07-30
Design Arena Web-app Agent: 1294 Elo2026-07-29
Design Arena Web-app Agent: 1292 Elo2026-07-31
Design Arena Web-app Agent: 1289 Elo2026-08-02
Design Arena Web-app Agent: 1291 Elo2026-08-01
Design Arena Web-app Agent: 1288 Elo2026-08-05
Design Arena Web-app Agent: 1280 Elo2026-08-09
Design Arena Web-app Agent: 1290 Elo2026-08-03
Design Arena Web-app Agent: 1290 Elo2026-08-04
Design Arena Web-app Agent: 1276 Elo2026-08-14
Design Arena Web-app Agent: 1287 Elo2026-08-06
Design Arena Web-app Agent: 1278 Elo2026-08-10
Design Arena Web-app Agent: 1287 Elo2026-08-07
Design Arena Web-app Agent: 1278 Elo2026-08-18
Design Arena Web-app Agent: 1287 Elo2026-08-08
Design Arena Web-app Agent: 1278 Elo2026-08-11
Design Arena Web-app Agent: 1276 Elo2026-08-15
Design Arena Web-app Agent: 1279 Elo2026-08-12
Design Arena Web-app Agent: 1277 Elo2026-08-13
Design Arena Web-app Agent: 1278 Elo2026-08-19
Design Arena Web-app Agent: 1269 Elo2026-08-27
Design Arena Web-app Agent: 1278 Elo2026-08-20
Design Arena Web-app Agent: 1282 Elo2026-08-16
Design Arena Web-app Agent: 1284 Elo2026-09-05
Design Arena Web-app Agent: 1282 Elo2026-08-17
Design Arena Web-app Agent: 1274 Elo2026-08-21
Design Arena Web-app Agent: 1274 Elo2026-08-22
Design Arena Web-app Agent: 1274 Elo2026-08-23
Design Arena Web-app Agent: 1274 Elo2026-08-28
Design Arena Web-app Agent: 1269 Elo2026-08-24
Design Arena Web-app Agent: 1269 Elo2026-08-25
Design Arena Web-app Agent: 1270 Elo2026-08-26
Design Arena Web-app Agent: 1284 Elo2026-09-06
Design Arena Web-app Agent: 1275 Elo2026-08-29
Design Arena Web-app Agent: 1279 Elo2026-08-30
Design Arena Web-app Agent: 1281 Elo2026-09-07
Design Arena Web-app Agent: 1283 Elo2026-08-31
Design Arena Web-app Agent: 1284 Elo2026-09-01
Design Arena Web-app Agent: 1284 Elo2026-09-02
Design Arena Web-app Agent: 1284 Elo2026-09-03
Design Arena Web-app Agent: 1282 Elo2026-09-08
Design Arena Web-app Agent: 1284 Elo2026-09-04
Design Arena Website: 1341 Elo2026-07-28
Design Arena Website: 1341 Elo2026-07-27
Design Arena Website: 1343 Elo2026-07-30
Design Arena Website: 1341 Elo2026-07-29
Design Arena Website: 1342 Elo2026-07-31
Design Arena Website: 1341 Elo2026-08-02
Design Arena Website: 1339 Elo2026-08-01
Design Arena Website: 1343 Elo2026-08-05
Design Arena Website: 1341 Elo2026-08-03
Design Arena Website: 1337 Elo2026-08-09
Design Arena Website: 1344 Elo2026-08-04
Design Arena Website: 1336 Elo2026-08-14
Design Arena Website: 1343 Elo2026-08-06
Design Arena Website: 1341 Elo2026-08-07
Design Arena Website: 1337 Elo2026-08-10
Design Arena Website: 1339 Elo2026-08-08
Design Arena Website: 1327 Elo2026-08-18
Design Arena Website: 1337 Elo2026-08-11
Design Arena Website: 1328 Elo2026-08-15
Design Arena Website: 1337 Elo2026-08-12
Design Arena Website: 1336 Elo2026-08-13
Design Arena Website: 1327 Elo2026-08-19
Design Arena Website: 1328 Elo2026-08-16
Design Arena Website: 1327 Elo2026-08-20
Design Arena Website: 1332 Elo2026-08-27
Design Arena Website: 1327 Elo2026-08-17
Design Arena Website: 1323 Elo2026-09-05
Design Arena Website: 1333 Elo2026-08-21
Design Arena Website: 1333 Elo2026-08-22
Design Arena Website: 1332 Elo2026-08-23
Design Arena Website: 1332 Elo2026-08-24
Design Arena Website: 1331 Elo2026-08-28
Design Arena Website: 1331 Elo2026-08-25
Design Arena Website: 1331 Elo2026-08-26
Design Arena Website: 1320 Elo2026-09-06
Design Arena Website: 1330 Elo2026-08-29
Design Arena Website: 1330 Elo2026-08-30
Design Arena Website: 1330 Elo2026-08-31
Design Arena Website: 1321 Elo2026-09-07
Design Arena Website: 1328 Elo2026-09-01
Design Arena Website: 1328 Elo2026-09-02
Design Arena Website: 1325 Elo2026-09-03
Design Arena Website: 1322 Elo2026-09-08
Design Arena Website: 1323 Elo2026-09-04
Frontier-Bench^‡: 51.8%2026-07-24Claude Code, max effort
LiveBench Agentic Coding: 65.2%2026-06-25
LiveBench Coding: 81.5%2026-06-25
LiveBench Data Analysis: 74.5%2026-06-25
LiveBench Instruction Following: 63.8%2026-06-25
LiveBench Language: 88.7%2026-06-25
LiveBench Math: 95.7%2026-06-25
LiveBench Reasoning: 91.2%2026-06-25
LMArena Agent: 11.92026-07-28
LMArena Agent: 12.12026-08-03
LMArena Agent: 122026-08-12
LMArena Agent: 122026-08-06
LMArena Agent: 122026-08-11
LMArena Agent: 12.22026-08-13
LMArena Agent: 12.72026-08-24
LMArena Agent: 13.82026-08-31
LMArena Agent: 12.72026-09-05
LMArena Agent: 132026-08-27
LMArena Creative Writing: 1511 Elo2026-07-27
LMArena Creative Writing: 1511 Elo2026-07-30
LMArena Creative Writing: 1498 Elo2026-08-02
LMArena Creative Writing: 1498 Elo2026-08-03
LMArena Creative Writing: 1493 Elo2026-08-06
LMArena Creative Writing: 1492 Elo2026-08-12
LMArena Creative Writing: 1486 Elo2026-08-21
LMArena Creative Writing: 1490 Elo2026-09-02
LMArena Creative Writing: 1487 Elo2026-08-27
LMArena Creative Writing: 1490 Elo2026-09-01
LMArena Instruction Following: 1521 Elo2026-07-27
LMArena Instruction Following: 1521 Elo2026-07-30
LMArena Instruction Following: 1515 Elo2026-08-02
LMArena Instruction Following: 1515 Elo2026-08-03
LMArena Instruction Following: 1514 Elo2026-08-06
LMArena Instruction Following: 1516 Elo2026-08-12
LMArena Instruction Following: 1514 Elo2026-08-21
LMArena Instruction Following: 1514 Elo2026-08-27
LMArena Instruction Following: 1516 Elo2026-09-01
LMArena Instruction Following: 1516 Elo2026-09-02
LMArena Long Query: 1519 Elo2026-07-27
LMArena Long Query: 1519 Elo2026-07-30
LMArena Long Query: 1516 Elo2026-08-02
LMArena Long Query: 1516 Elo2026-08-03
LMArena Long Query: 1517 Elo2026-08-06
LMArena Long Query: 1517 Elo2026-08-12
LMArena Long Query: 1517 Elo2026-08-21
LMArena Long Query: 1516 Elo2026-08-27
LMArena Long Query: 1516 Elo2026-09-01
LMArena Long Query: 1516 Elo2026-09-02
LMArena Math: 1544 Elo2026-07-27
LMArena Math: 1544 Elo2026-07-30
LMArena Math: 1564 Elo2026-08-02
LMArena Math: 1564 Elo2026-08-03
LMArena Math: 1567 Elo2026-08-06
LMArena Math: 1570 Elo2026-08-12
LMArena Math: 1559 Elo2026-08-21
LMArena Math: 1551 Elo2026-08-27
LMArena Math: 1543 Elo2026-09-02
LMArena Math: 1542 Elo2026-09-01
LMArena Text: 1495 Elo2026-07-27
LMArena Text: 1495 Elo2026-07-30
LMArena Text: 1492 Elo2026-08-02
LMArena Text: 1493 Elo2026-08-06
LMArena Text: 1492 Elo2026-08-03
LMArena Text: 1494 Elo2026-08-12
LMArena Text: 1494 Elo2026-08-10
LMArena Text: 1493 Elo2026-08-21
LMArena Text: 1492 Elo2026-09-01
LMArena Text: 1493 Elo2026-09-02
LMArena Text: 1492 Elo2026-08-27
LMArena Vision: 1299 Elo2026-07-26
LMArena Vision: 1299 Elo2026-07-30
LMArena Vision: 1295 Elo2026-08-02
LMArena Vision: 1297 Elo2026-08-06
LMArena Vision: 1292 Elo2026-08-21
LMArena Vision: 1290 Elo2026-08-27
LMArena WebDev: 1670 Elo2026-07-27
LMArena WebDev: 1712 Elo2026-07-28
LMArena WebDev: 1712 Elo2026-07-30
LMArena WebDev: 1705 Elo2026-08-01
LMArena WebDev: 1704 Elo2026-07-31
LMArena WebDev: 1686 Elo2026-08-06
LMArena WebDev: 1705 Elo2026-08-02
LMArena WebDev: 1692 Elo2026-08-10
LMArena WebDev: 1691 Elo2026-08-12
LMArena WebDev: 1692 Elo2026-08-14
LMArena WebDev: 1688 Elo2026-09-01
LMArena WebDev: 1688 Elo2026-09-05
LMArena WebDev: 1691 Elo2026-08-21
LMArena WebDev: 1688 Elo2026-08-30
OSWorld Verified^‡: 34.7%2026-06-24max effort
SWE-rebench^‡: 63.4% ± 1.42026-07-01tools
Terminal-Bench 2.1^‡: 51.8%2026-07-24max 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 Autonomous AgentsEstablished evidence
100% of intended weighted task evidence across 6 independent sources. 7 of 7 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 Computer & Browser UseEstablished evidence
90% of intended weighted task evidence across 2 independent sources. 4 of 4 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 LLM for AI AgentsEstablished evidence
70% of intended weighted task evidence across 5 independent sources. 7 of 7 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 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.
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. Official release evidence is also tracked provisionally.
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. Official release evidence is also tracked provisionally.
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.
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 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.

“Box found that Opus 5 outperforms Opus 4.8 by 8% and delivers notable performance gains in the data analysis (11% improvement) and due diligence (17% improvement) workflows”
Anthropic · Introducing Claude Opus 5 \ Anthropic
“Within Devin, it also shows particular strength on difficult debugging and root-cause analysis tasks.”
Anthropic · Introducing Claude Opus 5 \ Anthropic
“It isn't just better on our hardest agentic coding tasks, up 22% over Opus 4.7, it's steadier, with far less variance run to run.”
Anthropic · Introducing Claude Opus 5 \ Anthropic
“Claude Opus 5 is the strongest Opus model we've tested on our trading benchmark, and it gets there using roughly a seventh of the reasoning tokens and under half the latency of Opus 4.8.”
Anthropic · Introducing Claude Opus 5 \ Anthropic

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.

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