DeepSeek V4 Pro 0423

deepseek/deepseek-v4-pro-20260423

DeepSeek V4 Pro, provided by DeepSeek, processes text input and generates text output with a context window of up to 1,048,576 tokens. It supports tool usage and reasoning, and is priced at $

tool use
structured outputs
reasoning
open-weight

At a glance

providerdeepseek

releasedApr 23, 2026

context1.0M

input / M$1.04

Specifications

What the API gives you

Context window: 1.0M tokens
Max output: 384K tokens
Input price: $1.04 / 1M tokens
Output price: $2.08 / 1M tokens
Cache read: $0.087 / 1M tokens
Modalities: text → text
Knowledge cutoff: —
Released: Apr 23, 2026

Measured evidence

Public benchmark record

Design Arena Coding: 1275 Elo2026-07-28
Design Arena Coding: 1275 Elo2026-07-27
Design Arena Coding: 1274 Elo2026-07-30
Design Arena Coding: 1274 Elo2026-07-29
Design Arena Coding: 1274 Elo2026-07-31
Design Arena Coding: 1273 Elo2026-08-02
Design Arena Coding: 1272 Elo2026-08-05
Design Arena Coding: 1273 Elo2026-08-01
Design Arena Coding: 1271 Elo2026-08-09
Design Arena Coding: 1273 Elo2026-08-03
Design Arena Coding: 1273 Elo2026-08-04
Design Arena Coding: 1267 Elo2026-08-14
Design Arena Coding: 1272 Elo2026-08-06
Design Arena Coding: 1271 Elo2026-08-10
Design Arena Coding: 1272 Elo2026-08-07
Design Arena Coding: 1261 Elo2026-08-18
Design Arena Coding: 1271 Elo2026-08-08
Design Arena Coding: 1261 Elo2026-08-15
Design Arena Coding: 1270 Elo2026-08-11
Design Arena Coding: 1269 Elo2026-08-12
Design Arena Coding: 1267 Elo2026-08-13
Design Arena Coding: 1260 Elo2026-08-19
Design Arena Coding: 1266 Elo2026-08-27
Design Arena Coding: 1266 Elo2026-08-21
Design Arena Coding: 1260 Elo2026-08-20
Design Arena Coding: 1261 Elo2026-08-16
Design Arena Coding: 1261 Elo2026-08-17
Design Arena Coding: 1267 Elo2026-08-22
Design Arena Coding: 1267 Elo2026-08-23
Design Arena Coding: 1266 Elo2026-08-28
Design Arena Coding: 1267 Elo2026-08-24
Design Arena Coding: 1266 Elo2026-08-25
Design Arena Coding: 1267 Elo2026-08-26
Design Arena Coding: 1266 Elo2026-08-29
Design Arena Coding: 1265 Elo2026-08-30
Design Arena Coding: 1265 Elo2026-08-31
Design Arena Coding: 1265 Elo2026-09-01
Design Arena Coding: 1264 Elo2026-09-02
Design Arena Coding: 1262 Elo2026-09-03
Design Arena Coding: 1261 Elo2026-09-04
Design Arena Full-stack Agent: 948 Elo2026-07-28
Design Arena Full-stack Agent: 948 Elo2026-07-27
Design Arena Full-stack Agent: 948 Elo2026-07-30
Design Arena Full-stack Agent: 948 Elo2026-07-29
Design Arena Full-stack Agent: 948 Elo2026-07-31
Design Arena Full-stack Agent: 948 Elo2026-08-02
Design Arena Full-stack Agent: 948 Elo2026-08-01
Design Arena Full-stack Agent: 948 Elo2026-08-05
Design Arena Full-stack Agent: 948 Elo2026-08-09
Design Arena Full-stack Agent: 948 Elo2026-08-03
Design Arena Full-stack Agent: 948 Elo2026-08-04
Design Arena Full-stack Agent: 948 Elo2026-08-14
Design Arena Full-stack Agent: 948 Elo2026-08-06
Design Arena Full-stack Agent: 948 Elo2026-08-10
Design Arena Full-stack Agent: 948 Elo2026-08-07
Design Arena Full-stack Agent: 948 Elo2026-08-18
Design Arena Full-stack Agent: 948 Elo2026-08-08
Design Arena Full-stack Agent: 948 Elo2026-08-11
Design Arena Full-stack Agent: 948 Elo2026-08-15
Design Arena Full-stack Agent: 948 Elo2026-08-12
Design Arena Full-stack Agent: 948 Elo2026-08-13
Design Arena Full-stack Agent: 948 Elo2026-08-19
Design Arena Full-stack Agent: 948 Elo2026-08-27
Design Arena Full-stack Agent: 948 Elo2026-08-20
Design Arena Full-stack Agent: 948 Elo2026-08-16
Design Arena Full-stack Agent: 948 Elo2026-08-17
Design Arena Full-stack Agent: 948 Elo2026-08-21
Design Arena Full-stack Agent: 948 Elo2026-08-22
Design Arena Full-stack Agent: 948 Elo2026-08-23
Design Arena Full-stack Agent: 948 Elo2026-08-28
Design Arena Full-stack Agent: 948 Elo2026-08-24
Design Arena Full-stack Agent: 948 Elo2026-08-25
Design Arena Full-stack Agent: 948 Elo2026-08-26
Design Arena Full-stack Agent: 948 Elo2026-08-29
Design Arena Full-stack Agent: 948 Elo2026-08-30
Design Arena Full-stack Agent: 948 Elo2026-08-31
Design Arena Full-stack Agent: 948 Elo2026-09-01
Design Arena Full-stack Agent: 948 Elo2026-09-02
Design Arena Full-stack Agent: 948 Elo2026-09-03
Design Arena Full-stack Agent: 948 Elo2026-09-04
Design Arena UI: 1260 Elo2026-07-28
Design Arena UI: 1261 Elo2026-07-27
Design Arena UI: 1260 Elo2026-07-30
Design Arena UI: 1261 Elo2026-07-29
Design Arena UI: 1259 Elo2026-07-31
Design Arena UI: 1259 Elo2026-08-02
Design Arena UI: 1259 Elo2026-08-01
Design Arena UI: 1258 Elo2026-08-05
Design Arena UI: 1256 Elo2026-08-09
Design Arena UI: 1259 Elo2026-08-03
Design Arena UI: 1259 Elo2026-08-04
Design Arena UI: 1253 Elo2026-08-14
Design Arena UI: 1258 Elo2026-08-06
Design Arena UI: 1257 Elo2026-08-07
Design Arena UI: 1256 Elo2026-08-10
Design Arena UI: 1255 Elo2026-08-08
Design Arena UI: 1251 Elo2026-08-18
Design Arena UI: 1255 Elo2026-08-11
Design Arena UI: 1252 Elo2026-08-15
Design Arena UI: 1256 Elo2026-08-12
Design Arena UI: 1256 Elo2026-08-13
Design Arena UI: 1251 Elo2026-08-19
Design Arena UI: 1251 Elo2026-08-16
Design Arena UI: 1251 Elo2026-08-20
Design Arena UI: 1249 Elo2026-08-27
Design Arena UI: 1251 Elo2026-08-17
Design Arena UI: 1249 Elo2026-08-21
Design Arena UI: 1250 Elo2026-08-22
Design Arena UI: 1250 Elo2026-08-23
Design Arena UI: 1250 Elo2026-08-24
Design Arena UI: 1249 Elo2026-08-28
Design Arena UI: 1250 Elo2026-08-25
Design Arena UI: 1251 Elo2026-08-26
Design Arena UI: 1249 Elo2026-08-29
Design Arena UI: 1249 Elo2026-08-30
Design Arena UI: 1249 Elo2026-08-31
Design Arena UI: 1249 Elo2026-09-01
Design Arena UI: 1248 Elo2026-09-02
Design Arena UI: 1246 Elo2026-09-03
Design Arena UI: 1244 Elo2026-09-04
Design Arena Web-app Agent: 1002 Elo2026-07-28
Design Arena Web-app Agent: 1002 Elo2026-07-27
Design Arena Web-app Agent: 999 Elo2026-07-30
Design Arena Web-app Agent: 999 Elo2026-07-29
Design Arena Web-app Agent: 999 Elo2026-07-31
Design Arena Web-app Agent: 1000 Elo2026-08-02
Design Arena Web-app Agent: 999 Elo2026-08-01
Design Arena Web-app Agent: 1000 Elo2026-08-05
Design Arena Web-app Agent: 1000 Elo2026-08-09
Design Arena Web-app Agent: 1000 Elo2026-08-03
Design Arena Web-app Agent: 1000 Elo2026-08-04
Design Arena Web-app Agent: 1000 Elo2026-08-14
Design Arena Web-app Agent: 1000 Elo2026-08-06
Design Arena Web-app Agent: 1000 Elo2026-08-10
Design Arena Web-app Agent: 1000 Elo2026-08-07
Design Arena Web-app Agent: 1000 Elo2026-08-08
Design Arena Web-app Agent: 1000 Elo2026-08-18
Design Arena Web-app Agent: 1000 Elo2026-08-11
Design Arena Web-app Agent: 1000 Elo2026-08-15
Design Arena Web-app Agent: 1000 Elo2026-08-12
Design Arena Web-app Agent: 1000 Elo2026-08-13
Design Arena Web-app Agent: 1000 Elo2026-08-19
Design Arena Web-app Agent: 1000 Elo2026-08-27
Design Arena Web-app Agent: 1000 Elo2026-08-20
Design Arena Web-app Agent: 1000 Elo2026-08-16
Design Arena Web-app Agent: 1000 Elo2026-08-17
Design Arena Web-app Agent: 1000 Elo2026-08-21
Design Arena Web-app Agent: 1000 Elo2026-08-22
Design Arena Web-app Agent: 1000 Elo2026-08-23
Design Arena Web-app Agent: 1000 Elo2026-08-28
Design Arena Web-app Agent: 1000 Elo2026-08-24
Design Arena Web-app Agent: 1000 Elo2026-08-25
Design Arena Web-app Agent: 1000 Elo2026-08-26
Design Arena Web-app Agent: 1000 Elo2026-08-29
Design Arena Web-app Agent: 1000 Elo2026-08-30
Design Arena Web-app Agent: 1000 Elo2026-08-31
Design Arena Web-app Agent: 1000 Elo2026-09-01
Design Arena Web-app Agent: 1000 Elo2026-09-02
Design Arena Web-app Agent: 1000 Elo2026-09-03
Design Arena Web-app Agent: 1000 Elo2026-09-04
Design Arena Website: 1261 Elo2026-07-28
Design Arena Website: 1260 Elo2026-07-30
Design Arena Website: 1260 Elo2026-07-27
Design Arena Website: 1260 Elo2026-07-29
Design Arena Website: 1260 Elo2026-07-31
Design Arena Website: 1259 Elo2026-08-02
Design Arena Website: 1260 Elo2026-08-01
Design Arena Website: 1259 Elo2026-08-05
Design Arena Website: 1260 Elo2026-08-03
Design Arena Website: 1259 Elo2026-08-09
Design Arena Website: 1260 Elo2026-08-04
Design Arena Website: 1255 Elo2026-08-14
Design Arena Website: 1259 Elo2026-08-06
Design Arena Website: 1259 Elo2026-08-07
Design Arena Website: 1259 Elo2026-08-10
Design Arena Website: 1259 Elo2026-08-08
Design Arena Website: 1247 Elo2026-08-18
Design Arena Website: 1258 Elo2026-08-11
Design Arena Website: 1246 Elo2026-08-15
Design Arena Website: 1257 Elo2026-08-12
Design Arena Website: 1255 Elo2026-08-13
Design Arena Website: 1247 Elo2026-08-19
Design Arena Website: 1246 Elo2026-08-16
Design Arena Website: 1247 Elo2026-08-20
Design Arena Website: 1254 Elo2026-08-27
Design Arena Website: 1246 Elo2026-08-17
Design Arena Website: 1254 Elo2026-08-21
Design Arena Website: 1255 Elo2026-08-22
Design Arena Website: 1255 Elo2026-08-23
Design Arena Website: 1255 Elo2026-08-24
Design Arena Website: 1255 Elo2026-08-28
Design Arena Website: 1255 Elo2026-08-25
Design Arena Website: 1255 Elo2026-08-26
Design Arena Website: 1254 Elo2026-08-29
Design Arena Website: 1254 Elo2026-08-30
Design Arena Website: 1254 Elo2026-08-31
Design Arena Website: 1254 Elo2026-09-01
Design Arena Website: 1253 Elo2026-09-02
Design Arena Website: 1251 Elo2026-09-03
Design Arena Website: 1250 Elo2026-09-04
LiveBench Agentic Coding: 42.6%2026-06-25
LiveBench Coding: 70.0%2026-06-25
LiveBench Data Analysis: 74.5%2026-06-25
LiveBench Instruction Following: 62.4%2026-06-25
LiveBench Language: 78.1%2026-06-25
LiveBench Math: 90.7%2026-06-25
LiveBench Reasoning: 82.7%2026-06-25
LMArena Agent: -1.12026-07-13
LMArena Agent: -0.82026-07-20
LMArena Agent: -0.72026-07-27
LMArena Agent: -1.22026-07-21
LMArena Agent: -0.82026-07-28
LMArena Agent: -0.52026-08-03
LMArena Agent: 02026-08-12
LMArena Agent: -0.32026-08-06
LMArena Agent: -0.12026-08-11
LMArena Agent: 0.12026-08-13
LMArena Agent: -0.22026-08-24
LMArena Agent: -0.22026-08-27
LMArena Agent: 0.12026-08-31
LMArena Creative Writing: 1443 Elo2026-07-21
LMArena Creative Writing: 1443 Elo2026-07-27
LMArena Creative Writing: 1443 Elo2026-07-30
LMArena Creative Writing: 1446 Elo2026-08-02
LMArena Creative Writing: 1446 Elo2026-08-03
LMArena Creative Writing: 1445 Elo2026-08-06
LMArena Creative Writing: 1445 Elo2026-08-12
LMArena Creative Writing: 1444 Elo2026-08-21
LMArena Creative Writing: 1444 Elo2026-08-27
LMArena Creative Writing: 1444 Elo2026-09-01
LMArena Instruction Following: 1444 Elo2026-07-21
LMArena Instruction Following: 1445 Elo2026-07-27
LMArena Instruction Following: 1445 Elo2026-07-30
LMArena Instruction Following: 1446 Elo2026-08-02
LMArena Instruction Following: 1446 Elo2026-08-03
LMArena Instruction Following: 1444 Elo2026-08-06
LMArena Instruction Following: 1445 Elo2026-08-12
LMArena Instruction Following: 1445 Elo2026-08-21
LMArena Instruction Following: 1445 Elo2026-08-27
LMArena Instruction Following: 1445 Elo2026-09-01
LMArena Long Query: 1458 Elo2026-07-21
LMArena Long Query: 1458 Elo2026-07-27
LMArena Long Query: 1458 Elo2026-07-30
LMArena Long Query: 1459 Elo2026-08-02
LMArena Long Query: 1459 Elo2026-08-03
LMArena Long Query: 1457 Elo2026-08-06
LMArena Long Query: 1458 Elo2026-08-12
LMArena Long Query: 1458 Elo2026-08-21
LMArena Long Query: 1458 Elo2026-08-27
LMArena Long Query: 1458 Elo2026-09-01
LMArena Math: 1456 Elo2026-07-21
LMArena Math: 1457 Elo2026-07-27
LMArena Math: 1457 Elo2026-07-30
LMArena Math: 1456 Elo2026-08-02
LMArena Math: 1456 Elo2026-08-03
LMArena Math: 1456 Elo2026-08-06
LMArena Math: 1443 Elo2026-08-12
LMArena Math: 1443 Elo2026-08-21
LMArena Math: 1443 Elo2026-08-27
LMArena Math: 1442 Elo2026-09-01
LMArena Text: 1450 Elo2026-07-16
LMArena Text: 1457 Elo2026-07-21
LMArena Text: 1458 Elo2026-08-02
LMArena Text: 1457 Elo2026-07-27
LMArena Text: 1457 Elo2026-07-30
LMArena Text: 1458 Elo2026-08-10
LMArena Text: 1457 Elo2026-08-06
LMArena Text: 1458 Elo2026-08-03
LMArena Text: 1458 Elo2026-08-12
LMArena Text: 1458 Elo2026-08-21
LMArena Text: 1458 Elo2026-08-27
LMArena Text: 1458 Elo2026-09-01
LMArena WebDev: 1459 Elo2026-07-16
LMArena WebDev: 1459 Elo2026-07-21
LMArena WebDev: 1459 Elo2026-07-20
LMArena WebDev: 1464 Elo2026-07-27
LMArena WebDev: 1464 Elo2026-07-24
LMArena WebDev: 1464 Elo2026-07-28
LMArena WebDev: 1464 Elo2026-07-30
LMArena WebDev: 1464 Elo2026-08-01
LMArena WebDev: 1465 Elo2026-07-31
LMArena WebDev: 1464 Elo2026-08-06
LMArena WebDev: 1464 Elo2026-08-02
LMArena WebDev: 1445 Elo2026-08-12
LMArena WebDev: 1464 Elo2026-08-10
LMArena WebDev: 1446 Elo2026-08-14
LMArena WebDev: 1445 Elo2026-08-21
LMArena WebDev: 1445 Elo2026-08-30
LMArena WebDev: 1445 Elo2026-09-01
SWE-rebench^‡: 42.7% ± 1.32026-05-15tools
SWE-rebench^‡: 41.4% ± 1.02026-07-01tools

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.

Open-weight model on Hugging Face.

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 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 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 CodingEstablished evidence
74% of intended weighted task evidence across 4 independent sources. 8 of 8 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 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 Agentic CodingEstablished evidence
64% of intended weighted task evidence across 4 independent sources. 8 of 8 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 Autonomous AgentsProvisional evidence
49% of intended weighted task evidence across 3 independent sources; broader independent confirmation is still missing from Frontier-Bench or OSWorld Verified. 7 of 7 configured evaluation feeds currently have results.
Best LLM for Hermes AgentProvisional evidence
45% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.
Best LLM for OpenHandsProvisional evidence
45% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.
Best LLM for GooseProvisional evidence
45% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.
Best LLM for OpenClawProvisional evidence
45% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.
Best LLM for AI AgentsProvisional evidence
42% of intended weighted task evidence across 3 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or OSWorld Verified. 7 of 7 configured evaluation feeds currently have results.
Best LLM for Document SummarizationProvisional evidence
40% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 configured evaluation feeds currently have results.
Best LLM for SummarizationProvisional evidence
40% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 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 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.
Best LLM for Massive ContextProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or LongBench v2. 3 of 4 configured evaluation feeds currently have results.
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.
Best LLM for Image UnderstandingProvisional evidence
20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Vision or VLMEvalKit tasks. 3 of 4 configured evaluation feeds currently have results.
Best LLM for RAG & Long ContextProvisional evidence
18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 configured evaluation feeds currently have results.
Best LLM for OCR & DocumentsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
Best LLM for Vision & DocumentsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
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.
Best LLM for Charts & DiagramsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
Best LLM for Document ParsingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

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.