DeepSeek V3.2 vs Gemini 2.5 Flash

Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.

DeepSeek V3.2

DeepSeek · Sep 2025

68.4

The sparse-attention release that halved DeepSeek's already-lowest-in-class prices while keeping GPT-class quality.

Gemini 2.5 Flash

Google · Jun 2025

The 2025 fast-tier favorite: hybrid reasoning, full multimodality and a 1M context at $0.30/$2.50.

The verdict

DeepSeek V3.2 wins 2 of the 2 benchmarks these models share, against 0 for Gemini 2.5 Flash. DeepSeek V3.2 is about 2.7x cheaper per blended million tokens (3:1 input:output mix). Gemini 2.5 Flash also takes 1.0M of context versus 128K for DeepSeek V3.2. And DeepSeek V3.2 is open-weights (MIT), so it can be self-hosted — a structural advantage if data control or vendor independence matters.

Benchmark head-to-head 20

SWE-bench Verified
67.8%~48.9%
DeepSeek V3.2Gemini 2.5 Flash
GPQA Diamond
79.9%~78.3%
DeepSeek V3.2Gemini 2.5 Flash

Specs & pricing

DeepSeek V3.2Gemini 2.5 Flash
modhub Index68.4
Input price / 1M$0.28$0.3
Output price / 1M$0.42$2.5
Context window128K1.0M
Max output64K66K
Open weightsyes (MIT)no
Reasoning modelyesyes
Multimodal inputtexttext, image, audio, video
Knowledge cutoffJul 2025Jan 2025
ReleasedSep 2025Jun 2025
Example monthly cost*$3.43$6.75

* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.

Frequently asked questions

Which is better, DeepSeek V3.2 or Gemini 2.5 Flash?
DeepSeek V3.2 wins 2 of the 2 benchmarks these models share, against 0 for Gemini 2.5 Flash. DeepSeek V3.2 is about 2.7x cheaper per blended million tokens (3:1 input:output mix). Gemini 2.5 Flash also takes 1.0M of context versus 128K for DeepSeek V3.2. And DeepSeek V3.2 is open-weights (MIT), so it can be self-hosted — a structural advantage if data control or vendor independence matters.
Which is cheaper, DeepSeek V3.2 or Gemini 2.5 Flash?
DeepSeek V3.2 costs $0.28/$0.42 per million input/output tokens, while Gemini 2.5 Flash costs $0.3/$2.5. For a typical workload of 10M input and 1.5M output tokens per month, that's $3.43 versus $6.75.
Which model is better for coding, DeepSeek V3.2 or Gemini 2.5 Flash?
On SWE-bench Verified — the standard agentic-coding benchmark — DeepSeek V3.2 scores 67.8% versus ~48.9% for Gemini 2.5 Flash, making DeepSeek V3.2 the stronger pick for coding agents.

More comparisons