Home  /  Compare  /  DeepSeek V4 Pro (0813) vs Gemini 3.1 Pro Preview
Head to head

DeepSeek V4 Pro (0813) vs Gemini 3.1 Pro Preview.

Verdict by category, tale of the tape, shared benchmarks only, and cost at three workloads — all from the tracked data.

Verdict by category

CategoryDeepSeek V4 Pro (0813)Gemini 3.1 Pro PreviewWinner
League Score63.262.2DeepSeek V4 Pro (0813)
Coding (SWE-bench %)73.675.6Gemini 3.1 Pro Preview
Reasoning (GPQA %)90.194.3Gemini 3.1 Pro Preview
Context (tokens)10000001000000Draw
Speed (tok/s)99.4——
Price (blended $/1M)$1.98$4.50DeepSeek V4 Pro (0813) (cheaper)

Tale of the tape

SpecDeepSeek V4 Pro (0813)Gemini 3.1 Pro Preview
MakerDeepSeekGoogle DeepMind
ReleasedApr 24, 2026Feb 19, 2026
League Score6362
Context1,000,000 tokens1,000,000 tokens
Max output384,00065,536
$/1M input$1.32$2.00
$/1M output$3.96$12.00
WeightsOpenClosed
Speed99.4 t/s—

Head-to-head benchmarks (shared only)

BenchmarkDeepSeek V4 Pro (0813)Gemini 3.1 Pro Preview
SWE-bench Verified73.675.6
GPQA Diamond90.194.3
LMArena Elo14621500

Cost at three workloads

WorkloadDeepSeek V4 Pro (0813)Gemini 3.1 Pro Preview
Light — 1M in / 250K out$2.31$5.00
Standard — 10M in / 2.5M out$23.10$50.00
Long-context — 50M in / 5M out$85.80$160.00

The pick

Pick DeepSeek V4 Pro (0813) if you want the higher overall League Score; you want the lower running cost; you want downloadable weights / self-hosting.

Pick Gemini 3.1 Pro Preview if coding is the priority.

All figures from the model database — sources and dates on each model page. Cost your own workload or compare a different pair.

Frequently asked questions

Which is better overall, DeepSeek V4 Pro (0813) or Gemini 3.1 Pro Preview?

DeepSeek V4 Pro (0813) leads on overall League Score, 63 to 62 — a gap of 1 points on a composite of sourced benchmarks. That is overall form, not every category: check the per-category verdicts before choosing.

Which is cheaper, DeepSeek V4 Pro (0813) or Gemini 3.1 Pro Preview?

DeepSeek V4 Pro (0813) is cheaper: $1.98 vs $4.50 blended per 1M tokens (3:1 input/output mix) — about 2.3x less per token.

Which has the bigger context window, DeepSeek V4 Pro (0813) or Gemini 3.1 Pro Preview?

Neither — both take 1M of context.

Which is better for coding, DeepSeek V4 Pro (0813) or Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview scores higher on SWE-bench Verified, 75.6% vs 73.6% of real GitHub issues resolved.

Which is better at reasoning, DeepSeek V4 Pro (0813) or Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview leads on GPQA Diamond, 94.3% vs 90.1% on graduate-level science questions.

Are DeepSeek V4 Pro (0813) and Gemini 3.1 Pro Preview open source?

DeepSeek V4 Pro (0813) is open weights — it can be downloaded and self-hosted; Gemini 3.1 Pro Preview is closed weights — API access only.