Head to head

DeepSeek V4.1 Flash vs Gemini 3.8 Flash

Gemini 3.8 Flash scores higher on 2 of the 3 benchmarks both models publish. The widest gap is on Terminal-Bench 4.0, where DeepSeek V4.1 Flash leads by 11.5 points. DeepSeek V4.1 Flash has open weights, so unlike Gemini 3.8 Flash you can run it on your own GPU or deploy it on OpenLLM Buddy with no per-token fees.

DeepSeek V4.1 FlashDeepSeek · OpenLLM Buddy catalog79.8compositeGemini 3.8 FlashGoogle · Closed API70.8composite

Shared benchmarks

Raw published scores; the higher score on each row is highlighted. LMArena Elo is a human-preference vote and is not counted as a win.

  • GPQA Diamond
    90.995.3

    Pass@1, max reasoning effortGPQA Diamond; competitor figure reported by OpenAI

  • Humanity's Last Exam
    36.847.8

    Pass@1, no tools, max effort (text-only subset: 39.1)Humanity's Last Exam, no tools; AA independent run (High)

  • Terminal-Bench 4.0
    31.219.7

    Terminal-Bench 4.0, Pass@1, max effortTerminal-Bench 4.0; AA independent run (High)

  • LMArena Text Elo · not counted
    14761496

    LMArena text overall, 'DeepSeek-V4.1-Flash-20260910 (Max)', ±6, as of 2026-10-27LMArena Text, Overall; snapshot Oct 27, 2026; Gemini 3.8 Flash (High); +/-4

Should you pick DeepSeek V4.1 Flash or Gemini 3.8 Flash?

Pick DeepSeek V4.1 Flash if…

  • You care most about Terminal-Bench 4.0, where it scores at least 2 points higher.
  • You need to self-host: keep prompts and data on your own GPU, avoid per-token fees, or fine-tune the weights.
  • You need fast responses: about 217 tokens per second through APIs, against 135.
  • Cost matters: $1.20 per million output tokens against $3.75.
  • You want it running in minutes on a dedicated GPU through OpenLLM Buddy, in an uncensored build.

Pick Gemini 3.8 Flash if…

  • You care most about GPQA Diamond, Humanity's Last Exam, where it scores at least 2 points higher.
  • You want a managed API and would rather not run or pay for GPUs yourself.
  • People prefer its answers in blind chats: LMArena Elo 1496 against 1476.

Hardware and cost to run

DeepSeek V4.1 Flash

Even at 4-bit it needs about 381 GB (4× H200 (564 GB)), well beyond a home workstation, so most people run it on rented multi-GPU servers. Full BF16 precision needs about 1.3 TB (Multiple nodes (over 1.1 TB)).

On OpenLLM Buddy: DeepSeek-V4.1-Flash-UNCENSORED-FP8 runs on 4× H200 141GB as an FP8 build, with an OpenAI-compatible API and no per-token fees. Prices start at $280 USD for 11 hours.

Deploy DeepSeek V4.1 Flash
Full GPU memory table for DeepSeek V4.1 Flash

Gemini 3.8 Flash

Gemini 3.8 Flash is only available through Google's API, so it cannot be self-hosted; it costs $0.75 per million input tokens and $3.75 per million output tokens.

Category profile

Category index uses every test each model reports, not only the shared ones.

ReasoningCodingAgenticKnowledge
DeepSeek V4.1 FlashGemini 3.8 Flash
Reasoning75.986.7
Coding72.127.9
Agentic91.3—
Knowledge—97.6

Specs side by side

DeepSeek V4.1 FlashGemini 3.8 Flash
LabDeepSeekGoogle
WeightsOpenAPI only
Released2026-09-102026-09-02
Parameters552B total, 16B active (MoE)—
Context1M1M
LicenseMIT—
4-bit memory381 GB · 4× H200 (564 GB)—
API price$0.300 in / $1.20 out per 1M tokens$0.750 in / $3.75 out per 1M tokens
API speed217 tok/s135 tok/s

Related comparisons

Sources

ModelTestScoreSettingSource
DeepSeek V4.1 FlashGPQA Diamond90.9Pass@1, max reasoning effortDeepSeek-V4.1-Flash model card (Hugging Face)
Gemini 3.8 FlashGPQA Diamond95.3GPQA Diamond; competitor figure reported by OpenAIOpenAI GPT-6 Astra launch post (Academic table)
DeepSeek V4.1 FlashHumanity's Last Exam36.8Pass@1, no tools, max effort (text-only subset: 39.1)DeepSeek-V4.1-Flash model card (Hugging Face)
Gemini 3.8 FlashHumanity's Last Exam47.8Humanity's Last Exam, no tools; AA independent run (High)Artificial Analysis
DeepSeek V4.1 FlashTerminal-Bench 4.031.2Terminal-Bench 4.0, Pass@1, max effortDeepSeek-V4.1-Flash model card (Hugging Face)
Gemini 3.8 FlashTerminal-Bench 4.019.7Terminal-Bench 4.0; AA independent run (High)Artificial Analysis
DeepSeek V4.1 FlashLMArena Text Elo1476LMArena text overall, 'DeepSeek-V4.1-Flash-20260910 (Max)', ±6, as of 2026-10-27Arena (LMArena) text leaderboard
Gemini 3.8 FlashLMArena Text Elo1496LMArena Text, Overall; snapshot Oct 27, 2026; Gemini 3.8 Flash (High); +/-4LMArena (arena.ai) Text leaderboard

Scores are published results from model cards, lab reports, and public leaderboards, each linked to its source. Settings such as thinking mode or test version differ between sources and are shown next to each score. OpenLLM Buddy catalog models are uncensored fine-tunes; their scores are those of the official base model. Data collected 2026-10-30. How the index works