Google DeepMind

Gemini 3.5 Flash Lite

Tool callingStructured outputReasoningImage inputAudio inputVideo inputFile input
Gemini 3.5 Flash Lite launch post cover

Fast Gemini model balancing multimodal reasoning, tool use, and cost

Context
1.0M
Released
2mo ago
Input / 1M tokens
$0.30
Output / 1M tokens
$2.50
Cached input / 1M tokens
$0.03
Cache write / 1M tokens
$0.08

Sources:models.devOpenRouter

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Other published benchmarks

EvaluationReported score and source
ARC-AGI-2
10.3%
Third-party evaluator · gemini-3.5-flash-lite_high
Reasoning effort: high
External evaluation via Epoch AI
Original-source terms apply
ARC-AGI-2
1.5%
Third-party evaluator · gemini-3.5-flash-lite_low
Reasoning effort: low
External evaluation via Epoch AI
Original-source terms apply
ARC-AGI-2
5.3%
Third-party evaluator · gemini-3.5-flash-lite_medium
Reasoning effort: medium
External evaluation via Epoch AI
Original-source terms apply
ARC-AGI-2
0.8%
Third-party evaluator · gemini-3.5-flash-lite_minimal
Reasoning effort: minimal
External evaluation via Epoch AI
Original-source terms apply
FrontierMath-Tier-4-v2-Private
0.0%
Epoch AI · gemini-3.5-flash-lite_high
Reasoning effort: high
Epoch AI evaluation
FrontierMath-Tiers-1-3-v2-Private
26.0%
Epoch AI · gemini-3.5-flash-lite_high
Reasoning effort: high
Epoch AI evaluation
GPQA diamond
83.3%
Epoch AI · gemini-3.5-flash-lite_high
Reasoning effort: high
Epoch AI evaluation
GPQA diamond
75.8%
Epoch AI · gemini-3.5-flash-lite_low
Reasoning effort: low
Epoch AI evaluation
GPQA diamond
74.2%
Epoch AI · gemini-3.5-flash-lite_minimal
Reasoning effort: minimal
Epoch AI evaluation
OTIS Mock AIME 2024-2025
71.1%
Epoch AI · gemini-3.5-flash-lite_high
Reasoning effort: high
Epoch AI evaluation
OTIS Mock AIME 2024-2025
60.0%
Epoch AI · gemini-3.5-flash-lite_low
Reasoning effort: low
Epoch AI evaluation
OTIS Mock AIME 2024-2025
51.1%
Epoch AI · gemini-3.5-flash-lite_minimal
Reasoning effort: minimal
Epoch AI evaluation

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Specifications

Model family
Gemini
Type
Language model
Predecessor
—
Successor
—
Built on
Released
July 21, 2026
Model ID
google/gemini-3.5-flash-lite
Openness
Closed
License
—
Input modalities
text, image, video, file, audio
Output modalities
text
Modality
text+image+file+audio+video->text
Tokenizer
Gemini
Max output tokens
66K
Knowledge cutoff
2026-03
Hugging Face
—
Source freshness
Metadata synced Oct 8, 2026, 10:17 PM UTC
Additional source evidence
Source model
google/gemini-3.5-flash-lite
Source family
gemini-flash-lite
Last updated
2026-07-21
Knowledge cutoff
2026-03
Model context
1.0M
Model output limit
66K
Open weights
No
File attachments
Yes
Reasoning
Yes
Tool calling
Yes
Structured output
Yes
Temperature control
Yes

Published benchmarks

BenchmarkScoreEvaluation detailsEvidence
SWE-Bench Pro54.2 · resolve rate2026-07-21Source
Terminal-Bench54 · accuracyTerminus 2 · 2.1 · 2026-07-21Source
MLE-Bench39.2 · average position score2026-07-21Source
GDPval-AA1140 · Elov2 · 2026-07-21Source
OSWorld-Verified74 · success rate2026-07-21Source
CharXiv Reasoning74.5 · accuracyno tools · 2026-07-21Source
CharXiv Reasoning76.5 · accuracywith tools · 2026-07-21Source
GDM-MRCR72.2 · accuracy128k average, 8-needle · v2 · 2026-07-21Source
GDM-MRCR21.3 · accuracy1M pointwise, 8-needle · v2 · 2026-07-21Source
Complete source record (JSON)
{
  "id": "google/gemini-3.5-flash-lite",
  "name": "Gemini 3.5 Flash Lite",
  "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
  "family": "gemini-flash-lite",
  "last_updated": "2026-07-21",
  "knowledge": "2026-03",
  "attachment": true,
  "reasoning": true,
  "tool_call": true,
  "structured_output": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text",
      "image",
      "video",
      "audio",
      "pdf"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 1048576,
    "output": 65536
  },
  "benchmarks": [
    {
      "name": "SWE-Bench Pro",
      "score": 54.2,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "resolve rate",
      "date": "2026-07-21"
    },
    {
      "name": "Terminal-Bench",
      "score": 54,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "accuracy",
      "harness": "Terminus 2",
      "version": "2.1",
      "date": "2026-07-21"
    },
    {
      "name": "MLE-Bench",
      "score": 39.2,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "average position score",
      "date": "2026-07-21"
    },
    {
      "name": "GDPval-AA",
      "score": 1140,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "Elo",
      "version": "v2",
      "date": "2026-07-21"
    },
    {
      "name": "OSWorld-Verified",
      "score": 74,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "success rate",
      "date": "2026-07-21"
    },
    {
      "name": "CharXiv Reasoning",
      "score": 74.5,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "accuracy",
      "variant": "no tools",
      "date": "2026-07-21"
    },
    {
      "name": "CharXiv Reasoning",
      "score": 76.5,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "accuracy",
      "variant": "with tools",
      "date": "2026-07-21"
    },
    {
      "name": "GDM-MRCR",
      "score": 72.2,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "accuracy",
      "variant": "128k average, 8-needle",
      "version": "v2",
      "date": "2026-07-21"
    },
    {
      "name": "GDM-MRCR",
      "score": 21.3,
      "source": "https://deepmind.google/models/model-cards/gemini-3-5-flash-lite/",
      "metric": "accuracy",
      "variant": "1M pointwise, 8-needle",
      "version": "v2",
      "date": "2026-07-21"
    }
  ],
  "release_date": "2026-07-21",
  "open_weights": false
}

Infrastructure

Provider availability

All providers
ProviderEndpointContextInput /M tokensOutput /M tokensCached input /M tokensCache write /M tokensUptime 1dData handling
Googlegoogle-vertex/global/flex1.0M$0.15$1.25$0.02$0.04100.00%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/global/flexBatch pricing1.0M$0.15$1.25$0.02$0.04100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Google AI Studiogoogle-ai-studio/flex1.0M$0.15$1.25$0.02$0.04100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Google AI Studiogoogle-ai-studio/flexBatch pricing1.0M$0.15$1.25$0.02$0.04100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/global1.0M$0.30$2.50$0.03$0.08100.00%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/globalBatch pricing1.0M$0.30$2.50$0.03$0.08100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Google AI Studiogoogle-ai-studio1.0M$0.30$2.50$0.03$0.08100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Google AI Studiogoogle-ai-studioBatch pricing1.0M$0.30$2.50$0.03$0.08100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/eu1.0M$0.33$2.75$0.03$0.08100.00%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/us1.0M$0.33$2.75$0.03$0.08100.00%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/euBatch pricing1.0M$0.33$2.75$0.03$0.08100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/usBatch pricing1.0M$0.33$2.75$0.03$0.08100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/global/priority1.0M$0.54$4.50$0.05$0.15100.00%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Googlegoogle-vertex/global/priorityBatch pricing1.0M$0.54$4.50$0.05$0.15100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Google AI Studiogoogle-ai-studio/priority1.0M$0.54$4.50$0.05$0.15100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Google AI Studiogoogle-ai-studio/priorityBatch pricing1.0M$0.54$4.50$0.05$0.15100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Provider data:OpenRouter

Text provenance

Gemini app/web

Google SynthID text watermarking

Google DeepMind says text generated in the Gemini app and web experience is watermarked with SynthID by adjusting token probabilities.

Official explanation