gemini-3-pro-preview

Explore gemini-3-pro-preview: model specifications, capabilities, benchmark results, platform pricing and original sources.

vertexAI/publishers/google/models/gemini-3-pro-preview@default

Specifications & capabilities

Published model ID
gemini-3-pro-preview@default

Evaluations

1–20 / 23

BFCL

90.65%

source id
bfcl
leaderboard name
BFCL Non-Live AST
dimension
toolUse
metric
Non-live AST accuracy
rank
1
Original result ↗

HELM Capabilities

90.3%

source id
helm
leaderboard name
HELM MMLU-Pro
dimension
intelligence
metric
chain_of_thought_correctness
rank
1
sample size
1000
source model id
google/gemini-3-pro-preview
Original result ↗

HELM Capabilities

80.269%

source id
helm
leaderboard name
HELM GPQA
dimension
intelligence
metric
chain_of_thought_correctness
rank
1
sample size
446
source model id
google/gemini-3-pro-preview
Original result ↗

HELM Capabilities

85.872%

source id
helm
leaderboard name
HELM WildBench
dimension
agentic
metric
wildbench_score_rescaled
rank
5
sample size
1000
source model id
google/gemini-3-pro-preview
Original result ↗

BFCL

63.12%

source id
bfcl
leaderboard name
BFCL Multi Turn
dimension
agentic
metric
Multi-turn accuracy
rank
6
Original result ↗

BFCL

80%

source id
bfcl
leaderboard name
BFCL Web Search
dimension
agentic
metric
Web-search accuracy
rank
7
Original result ↗

BFCL

60.75%

source id
bfcl
leaderboard name
BFCL Multi Turn
dimension
agentic
metric
Multi-turn accuracy
rank
10
Original result ↗

HELM Capabilities

87.646%

source id
helm
leaderboard name
HELM IFEval
dimension
agentic
metric
ifeval_strict_accuracy
rank
13
sample size
541
source model id
google/gemini-3-pro-preview
Original result ↗

HELM Capabilities

55.55%

source id
helm
leaderboard name
HELM Omni-MATH
dimension
intelligence
metric
omni_math_accuracy
rank
13
sample size
1000
source model id
google/gemini-3-pro-preview
Original result ↗

BFCL

68.5%

source id
bfcl
leaderboard name
BFCL Web Search
dimension
agentic
metric
Web-search accuracy
rank
17
Original result ↗

SWE-bench

74.2%

source id
sweBench
leaderboard name
SWE-bench Verified
dimension
coding
metric
Resolved tasks
rank
21
sample size
500
evaluated at
2025-11-18T00:00:00.000Z
Original result ↗

BFCL

85.59%

source id
bfcl
leaderboard name
BFCL Irrelevance Detection
dimension
toolUse
metric
Irrelevance detection
rank
31
Original result ↗

BFCL

85.75%

source id
bfcl
leaderboard name
BFCL Non-Live AST
dimension
toolUse
metric
Non-live AST accuracy
rank
38
Original result ↗

SWE-bench

69.6%

source id
sweBench
leaderboard name
SWE-bench Verified
dimension
coding
metric
Resolved tasks
rank
48
sample size
500
evaluated at
2026-02-26T00:00:00.000Z
Original result ↗

Provider specifications & access

1–1 / 1

gemini-3-pro-preview ↗

source model id
publishers/google/models/gemini-3-pro-preview@default
hosting
vertex-managed
platforms
  • Google Cloud Vertex AI
architecture
input_modalities
    output_modalities
      capabilities

        Safety assessments

        1–1 / 1

        Foundation Model Transparency Index 2025 ↗

        41 / 100

        source id
        stanfordFMTI
        canonical model id
        vertexAI/publishers/google/models/gemini-3-pro-preview@default
        taxonomy
        transparency
        hazard category
        transparency
        hazard name
        Organization transparency
        policy area
        Google
        assessment type
        organization-level disclosure
        metric
        disclosure_score
        result label
        41 / 100
        sample size
        100
        updated at
        2026-11-09T18:08:05.227Z

        Original sources

        1–9 / 9

        Google DeepMind · Model reference ↗

        Google · Model reference ↗

        Google Cloud · Model page ↗

        Epoch AI · Catalog ↗

        GitHub · Catalog ↗

        Google Cloud · Catalog ↗

        crfm.stanford.edu · Catalog ↗

        gorilla.cs.berkeley.edu · Catalog ↗

        www.swebench.com · Catalog ↗

        Source links

        1–6 / 6

        Explore more models

        Models

        Loading
        EVIDENCE / COMPARISON

        Benchmarks

        Explore every benchmark

        Loading

        These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

        Install TensorTrace