Gemini 2.5 Flash

google/gemini-2.5-flash

Google's Gemini 2.5 Flash, released on 2025-06-17, accepts file, image, text, audio, and video inputs and generates text output. It supports tool usage and reasoning, features a context length of 1,048,576 tokens, and is priced at $0.3 per million input tokens and $2.5 per million output tokens. The model is proprietary and does not have open weights.

  • tool use
  • structured outputs
  • reasoning

At a glance

providergoogle
releasedMar 20, 2025
context1.0M
input / M$0.30

Specifications

What the API gives you

Context window
1.0M tokens
Max output
66K tokens
Input price
$0.30 / 1M tokens
Output price
$2.50 / 1M tokens
Cache read
$0.030 / 1M tokens
Modalities
file, image, text, audio, video → text
Knowledge cutoff
Jan 31, 2025
Released
Mar 20, 2025

Measured evidence

Public benchmark record

BFCL
56.2%2026-07-28
BFCL
56.2%2026-07-27
BFCL
56.2%2026-07-29
BFCL
56.2%2026-07-31
BFCL
56.2%2026-07-30
BFCL
56.2%2026-08-02
BFCL
56.2%2026-08-01
BFCL
56.2%2026-08-05
BFCL
56.2%2026-08-03
BFCL
56.2%2026-08-09
BFCL
56.2%2026-08-04
BFCL
56.2%2026-08-06
BFCL
56.2%2026-08-07
BFCL
56.2%2026-08-14
BFCL
56.2%2026-08-10
BFCL
56.2%2026-08-08
BFCL
56.2%2026-08-18
BFCL
56.2%2026-08-11
BFCL
56.2%2026-08-15
BFCL
56.2%2026-08-12
BFCL
56.2%2026-08-13
BFCL
56.2%2026-08-19
BFCL
56.2%2026-08-16
BFCL
56.2%2026-08-20
BFCL
56.2%2026-08-27
BFCL
56.2%2026-08-17
BFCL
56.2%2026-08-21
BFCL
56.2%2026-08-22
BFCL
56.2%2026-08-23
BFCL
56.2%2026-08-24
BFCL
56.2%2026-09-05
BFCL
56.2%2026-08-28
BFCL
56.2%2026-08-25
BFCL
56.2%2026-08-26
BFCL
56.2%2026-09-06
BFCL
56.2%2026-08-29
BFCL
56.2%2026-08-30
BFCL
56.2%2026-08-31
BFCL
56.2%2026-09-01
BFCL
56.2%2026-09-02
BFCL
56.2%2026-09-03
BFCL
56.2%2026-09-04
Design Arena Coding
1090 Elo2026-07-28
Design Arena Coding
1090 Elo2026-07-27
Design Arena Coding
1090 Elo2026-07-30
Design Arena Coding
1090 Elo2026-07-29
Design Arena Coding
1089 Elo2026-07-31
Design Arena Coding
1089 Elo2026-08-02
Design Arena Coding
1088 Elo2026-08-05
Design Arena Coding
1089 Elo2026-08-01
Design Arena Coding
1087 Elo2026-08-09
Design Arena Coding
1089 Elo2026-08-03
Design Arena Coding
1088 Elo2026-08-04
Design Arena Coding
1084 Elo2026-08-14
Design Arena Coding
1088 Elo2026-08-06
Design Arena Coding
1087 Elo2026-08-10
Design Arena Coding
1088 Elo2026-08-07
Design Arena Coding
1078 Elo2026-08-18
Design Arena Coding
1087 Elo2026-08-08
Design Arena Coding
1078 Elo2026-08-15
Design Arena Coding
1087 Elo2026-08-11
Design Arena Coding
1086 Elo2026-08-12
Design Arena Coding
1084 Elo2026-08-13
Design Arena Coding
1077 Elo2026-08-19
Design Arena Coding
1083 Elo2026-08-27
Design Arena Coding
1077 Elo2026-08-20
Design Arena Coding
1078 Elo2026-08-16
Design Arena Coding
1076 Elo2026-09-05
Design Arena Coding
1078 Elo2026-08-17
Design Arena Coding
1084 Elo2026-08-21
Design Arena Coding
1084 Elo2026-08-22
Design Arena Coding
1084 Elo2026-08-23
Design Arena Coding
1083 Elo2026-08-28
Design Arena Coding
1084 Elo2026-08-24
Design Arena Coding
1084 Elo2026-08-25
Design Arena Coding
1084 Elo2026-08-26
Design Arena Coding
1075 Elo2026-09-06
Design Arena Coding
1083 Elo2026-08-29
Design Arena Coding
1083 Elo2026-08-30
Design Arena Coding
1083 Elo2026-08-31
Design Arena Coding
1083 Elo2026-09-01
Design Arena Coding
1082 Elo2026-09-02
Design Arena Coding
1080 Elo2026-09-03
Design Arena Coding
1078 Elo2026-09-04
Design Arena UI
1040 Elo2026-07-28
Design Arena UI
1041 Elo2026-07-27
Design Arena UI
1041 Elo2026-07-30
Design Arena UI
1041 Elo2026-07-29
Design Arena UI
1040 Elo2026-07-31
Design Arena UI
1041 Elo2026-08-02
Design Arena UI
1041 Elo2026-08-01
Design Arena UI
1041 Elo2026-08-05
Design Arena UI
1038 Elo2026-08-09
Design Arena UI
1041 Elo2026-08-03
Design Arena UI
1041 Elo2026-08-04
Design Arena UI
1036 Elo2026-08-14
Design Arena UI
1040 Elo2026-08-06
Design Arena UI
1040 Elo2026-08-07
Design Arena UI
1038 Elo2026-08-10
Design Arena UI
1038 Elo2026-08-08
Design Arena UI
1033 Elo2026-08-18
Design Arena UI
1038 Elo2026-08-11
Design Arena UI
1034 Elo2026-08-15
Design Arena UI
1039 Elo2026-08-12
Design Arena UI
1039 Elo2026-08-13
Design Arena UI
1033 Elo2026-08-19
Design Arena UI
1033 Elo2026-08-16
Design Arena UI
1033 Elo2026-08-20
Design Arena UI
1031 Elo2026-08-27
Design Arena UI
1033 Elo2026-08-17
Design Arena UI
1027 Elo2026-09-05
Design Arena UI
1032 Elo2026-08-21
Design Arena UI
1032 Elo2026-08-22
Design Arena UI
1032 Elo2026-08-23
Design Arena UI
1032 Elo2026-08-24
Design Arena UI
1031 Elo2026-08-28
Design Arena UI
1032 Elo2026-08-25
Design Arena UI
1033 Elo2026-08-26
Design Arena UI
1028 Elo2026-09-06
Design Arena UI
1031 Elo2026-08-29
Design Arena UI
1032 Elo2026-08-30
Design Arena UI
1031 Elo2026-08-31
Design Arena UI
1031 Elo2026-09-01
Design Arena UI
1031 Elo2026-09-02
Design Arena UI
1029 Elo2026-09-03
Design Arena UI
1027 Elo2026-09-04
Design Arena Website
1099 Elo2026-07-28
Design Arena Website
1099 Elo2026-07-30
Design Arena Website
1099 Elo2026-07-27
Design Arena Website
1099 Elo2026-07-29
Design Arena Website
1099 Elo2026-07-31
Design Arena Website
1098 Elo2026-08-02
Design Arena Website
1098 Elo2026-08-01
Design Arena Website
1098 Elo2026-08-05
Design Arena Website
1098 Elo2026-08-03
Design Arena Website
1097 Elo2026-08-09
Design Arena Website
1098 Elo2026-08-04
Design Arena Website
1094 Elo2026-08-14
Design Arena Website
1098 Elo2026-08-06
Design Arena Website
1098 Elo2026-08-07
Design Arena Website
1097 Elo2026-08-10
Design Arena Website
1097 Elo2026-08-08
Design Arena Website
1085 Elo2026-08-18
Design Arena Website
1097 Elo2026-08-11
Design Arena Website
1085 Elo2026-08-15
Design Arena Website
1096 Elo2026-08-12
Design Arena Website
1095 Elo2026-08-13
Design Arena Website
1085 Elo2026-08-19
Design Arena Website
1085 Elo2026-08-16
Design Arena Website
1085 Elo2026-08-20
Design Arena Website
1094 Elo2026-08-27
Design Arena Website
1085 Elo2026-08-17
Design Arena Website
1090 Elo2026-09-05
Design Arena Website
1094 Elo2026-08-21
Design Arena Website
1094 Elo2026-08-22
Design Arena Website
1094 Elo2026-08-23
Design Arena Website
1094 Elo2026-08-24
Design Arena Website
1094 Elo2026-08-28
Design Arena Website
1094 Elo2026-08-25
Design Arena Website
1094 Elo2026-08-26
Design Arena Website
1087 Elo2026-09-06
Design Arena Website
1094 Elo2026-08-29
Design Arena Website
1093 Elo2026-08-30
Design Arena Website
1093 Elo2026-08-31
Design Arena Website
1093 Elo2026-09-01
Design Arena Website
1092 Elo2026-09-02
Design Arena Website
1090 Elo2026-09-03
Design Arena Website
1090 Elo2026-09-04
LMArena Creative Writing
1402 Elo2026-07-21
LMArena Creative Writing
1402 Elo2026-07-27
LMArena Creative Writing
1402 Elo2026-07-30
LMArena Creative Writing
1402 Elo2026-08-02
LMArena Creative Writing
1402 Elo2026-08-03
LMArena Creative Writing
1402 Elo2026-08-06
LMArena Creative Writing
1402 Elo2026-08-12
LMArena Creative Writing
1401 Elo2026-08-21
LMArena Creative Writing
1402 Elo2026-09-02
LMArena Creative Writing
1402 Elo2026-08-27
LMArena Creative Writing
1402 Elo2026-09-01
LMArena Long Query
1420 Elo2026-07-21
LMArena Long Query
1420 Elo2026-07-27
LMArena Long Query
1420 Elo2026-08-06
LMArena Long Query
1420 Elo2026-07-30
LMArena Long Query
1420 Elo2026-08-02
LMArena Long Query
1420 Elo2026-08-03
LMArena Long Query
1420 Elo2026-08-12
LMArena Long Query
1420 Elo2026-08-21
LMArena Long Query
1419 Elo2026-09-02
LMArena Long Query
1419 Elo2026-08-27
LMArena Long Query
1419 Elo2026-09-01
LMArena Math
1410 Elo2026-07-21
LMArena Math
1410 Elo2026-07-27
LMArena Math
1410 Elo2026-07-30
LMArena Math
1410 Elo2026-08-02
LMArena Math
1410 Elo2026-08-03
LMArena Math
1410 Elo2026-08-06
LMArena Math
1410 Elo2026-08-12
LMArena Math
1410 Elo2026-08-21
LMArena Math
1411 Elo2026-09-01
LMArena Math
1411 Elo2026-08-27
LMArena Text
1417 Elo2026-07-16
LMArena Text
1410 Elo2026-07-21
LMArena Text
1410 Elo2026-07-27
LMArena Text
1410 Elo2026-07-30
LMArena Text
1410 Elo2026-08-10
LMArena Text
1410 Elo2026-08-02
LMArena Text
1410 Elo2026-08-06
LMArena Text
1410 Elo2026-08-03
LMArena Text
1410 Elo2026-08-21
LMArena Text
1410 Elo2026-08-12
LMArena Text
1410 Elo2026-09-01
LMArena Text
1410 Elo2026-09-02
LMArena Text
1410 Elo2026-08-27
LMArena Vision
1235 Elo2026-07-12
LMArena Vision
1214 Elo2026-07-21
LMArena Vision
1214 Elo2026-07-26
LMArena Vision
1214 Elo2026-07-30
LMArena Vision
1214 Elo2026-08-02
LMArena Vision
1214 Elo2026-08-06
LMArena Vision
1214 Elo2026-08-21
LMArena Vision
1214 Elo2026-08-27
SWE-rebench
19.9% ± 0.42025-11-01tools

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.

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 Business WritingProvisional evidence

    70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Marketing CopyProvisional evidence

    70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Chat & RoleplayProvisional evidence

    70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • 70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • 65% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or LMArena Search. 4 of 5 configured evaluation feeds currently have results.

  • Best LLM for Creative WritingProvisional evidence

    60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for WritingProvisional evidence

    60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Frontend & UIProvisional evidence

    60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Coding or LMArena WebDev. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Hermes AgentProvisional evidence

    55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LMArena Agent or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for OpenHandsProvisional evidence

    55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LMArena Agent or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for GooseProvisional evidence

    55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LMArena Agent or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for OpenClawProvisional evidence

    55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LMArena Agent or Terminal-Bench 2.1. 5 of 5 configured evaluation feeds currently have results.

  • 55% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Reasoning or VLMEvalKit tasks. 3 of 4 configured evaluation feeds currently have results.

  • 55% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • 50% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for AI AgentsProvisional evidence

    50% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LiveBench Agentic Coding or LMArena Agent. 7 of 7 configured evaluation feeds currently have results.

  • 40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 6 of 6 configured evaluation feeds currently have results.

  • Best LLM for CodingProvisional evidence

    40% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 8 of 8 configured evaluation feeds currently have results.

  • Best LLM for Competition MathProvisional evidence

    40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Math & ReasoningProvisional evidence

    40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best Value LLMProvisional evidence

    36% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • 33% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.

  • 33% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.

  • 33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Autonomous AgentsProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Frontier-Bench or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Fast SummariesProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • 25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • Best Cheap LLM for High VolumeProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • 25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • Cheapest Capable LLMProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for TranslationProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 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 OCR & DocumentsProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or Structured-output eval. 3 of 5 configured evaluation feeds currently have results.

  • 25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or Structured-output eval. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for Charts & DiagramsProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or Structured-output eval. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for Document ParsingProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or Structured-output eval. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for Agentic CodingProvisional evidence

    23% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Design Arena Full-stack or Design Arena Web-apps. 8 of 8 configured evaluation feeds currently have results.

  • 20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.

  • Best LLM for SummarizationProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.

  • 20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or OSWorld Verified. 4 of 4 configured evaluation feeds currently have results.

  • Best Local LLM for CodingProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 5 of 5 configured evaluation feeds currently have results.

  • 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.

  • 13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Route reliability. 2 of 4 configured evaluation feeds currently have results.

  • Best LLM for Code CompletionProvisional evidence

    13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 4 of 4 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.

Recommendation coverage

Where this model appears

  1. #4Best LLM for Document Parsing

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