Sign Up

Decision support

Gemini 3.8 Flash vs Pareto 26.10

Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.

ModelsHardware
Gemini 3.8 Flash

Google DeepMind

Sources:models.devOpenRouter

Pareto 26.10

Unbiased

Sources:OpenRouter

Benchmark results

  • DeepSWE 1.1

    Best pass@1 · Higher is better

    • Gemini 3.8 Flash
      73.8%
      #2 of 28
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • medium71.0%mini-swe-agent · pass@4 83.2% · $1.97 / attempt
    • high73.8%mini-swe-agent · pass@4 85.8% · $2.36 / attempt
    Pareto 26.10

    Not reported

    Source:DeepSWE leaderboard, 2026-09-22Official source JSON

  • Arena Text

    Arena rating · Higher is better

    • Gemini 3.8 Flash
      1497
      Arena #9CI 1493 to 1502
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high149795% CI 1493 to 1502 · 61,143 votes
    Pareto 26.10

    Not reported

    Source:Arena Text leaderboard, 2026-11-09

  • Arena WebDev

    Arena rating · Higher is better

    • Gemini 3.8 Flash
      1583
      Arena #42CI 1576 to 1589
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high158395% CI 1576 to 1589 · 13,847 votes
    Pareto 26.10

    Not reported

    Source:Arena WebDev leaderboard, 2026-11-10

  • Arena Vision

    Arena rating · Higher is better

    • Gemini 3.8 Flash
      1293
      Arena #11CI 1285 to 1301
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high129395% CI 1285 to 1301 · 7,624 votes
    Pareto 26.10

    Not reported

    Source:Arena Vision leaderboard, 2026-11-09

  • Arena Document

    Arena rating · Higher is better

    • Gemini 3.8 Flash
      1469
      Arena #19CI 1461 to 1477
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high146995% CI 1461 to 1477 · 6,950 votes
    Pareto 26.10

    Not reported

    Source:Arena Document leaderboard, 2026-11-09

  • Agent Arena

    IPS score · Higher is better

    • Gemini 3.8 Flash
      +0.023
      Arena #20CI +0.014 to +0.032
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high+0.02395% CI +0.014 to +0.032 · 8,602,306 observations
    Pareto 26.10

    Not reported

    Source:Agent Arena leaderboard, 2026-11-11

  • ARC-AGI-2

    Best published score · Higher is better

    • Gemini 3.8 Flash
      89.2%
      #10 of 85
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • low77.5%
    • medium82.9%
    • high89.2%
    Pareto 26.10

    Not reported

    Source:External evaluation via Epoch AIOriginal-source terms apply

  • FrontierMath-Tier-4-v2-Private

    Best published score · Higher is better

    • Gemini 3.8 Flash
      22.0%
      #47 of 72
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high22.0%
    Pareto 26.10

    Not reported

    Source:Epoch AI evaluation

  • FrontierMath-Tiers-1-3-v2-Private

    Best published score · Higher is better

    • Gemini 3.8 Flash
      68.4%
      #31 of 87
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high68.4%
    Pareto 26.10

    Not reported

    Source:Epoch AI evaluation

  • GPQA diamond

    Best published score · Higher is better

    • Gemini 3.8 Flash
      95.4%
      #5 of 161
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high95.4%
    Pareto 26.10

    Not reported

    Source:Epoch AI evaluation

  • Humanity's Last Exam

    Best published score · Higher is better

    • Gemini 3.8 Flash
      44.5%
      #5 of 34
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • unknown44.5%
    Pareto 26.10

    Not reported

    Source:External evaluation via Epoch AIOriginal-source terms apply

  • OTIS Mock AIME 2024-2025

    Best published score · Higher is better

    • Gemini 3.8 Flash
      98.9%
      #19 of 158
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high98.9%
    Pareto 26.10

    Not reported

    Source:Epoch AI evaluation

  • SimpleQA Verified

    Best published score · Higher is better

    • Gemini 3.8 Flash
      69.7%
      #10 of 87
    • Pareto 26.10
      Not reported
    Runs and sources
    Gemini 3.8 Flash
    • high69.7%
    Pareto 26.10

    Not reported

    Source:Epoch AI evaluation

Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.

Original-source terms apply to evaluations credited to their original publishers.

DeepSWE 1.1 results: official leaderboard, data generated 2026-09-22. Only exact model versions in the official dataset are matched.

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.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

Model specifications

Release date
September 2, 2026
October 15, 2026
Context window
1M
1M
Native context
—
—
Max output tokens
65,536
131,072
Knowledge cutoff
—
—
Open weights
Closed
Unknown
Parameter count
—
—

API pricing · USD per 1M tokens

Input
$0.825
$0.80
Output
$4.125
$3.20
Cached input
$0.0825
$0.03
Cache write
$0.0417
—

Capabilities and modalities

Input modalities
text, image, video, file, audio
text, image
Output modalities
text
text
Tool calling
Supported
Supported
Reasoning option
Supported
—
Supported options
include_reasoning, max_tokens, reasoning, reasoning_effort, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p
max_tokens, temperature, tool_choice, tools, top_p
Capabilities
—
—

Architecture and availability

Architecture
—
—
Model families
—
—
Tokenizer
Gemini
Other
Instruction format
—
—
License
—
—
Providers
2
1
Provider endpoints
14
1
Ollama variants
0
0
Metadata source
ai-research
ai-research

A dash means the field is not reported. Provider-specific performance and pricing may differ.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.