DeepSeek V4 Flash vs Gemini 2.5 Flash
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by DeepSeek
3 providers:
DeepSeek $0.140/$0.280
Fireworks $0.140/$0.280
DeepInfra $0.090/$0.180
|
by Google
|
|---|---|---|
| Overview | ||
| Status | Current budget | Deprecated budget |
| Released | Apr 24, 2026 | Jun 17, 2025 |
| Pricing per million tokens | ||
| Input | $0.140/Mtok | $0.300/Mtok |
| Output | $0.280/Mtok | $2.50/Mtok |
| Blended avg | $0.175/Mtok | $0.850/Mtok |
| Cached input | $0.003/Mtok | — |
| Price basis | Correct as published today. DeepSeek has announced that from 16:00 UTC on 2026-08-16 this model moves to peak/off-peak billing: peak $0.44 input / $0.014 cached input / $1.32 output per 1M tokens, off-peak exactly half ($0.22 / $0.007 / $0.66). Peak hours are 01:00-04:00 and 06:00-10:00 UTC; every other hour is off-peak. This row tracks the flat rate DeepSeek charges until that moment, and switches to the peak rate when it takes effect. | — |
| Specifications | ||
| Context window | 1M tokens | 1M tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
text
|
textimageaudiovideo
|
| Benchmarks source: Vellum LLM Leaderboard | ||
| Avg benchmark score | 77.5 | 50.2 |
| Perf / dollar | 442.9 | 59.1 |
| GPQA Diamond | 88.1 | 78.3 |
| SWE-Bench Verified | 79 vendor | 38 |
| LiveCodeBench | 91.6 | 63.5 |
| Humanity's Last Exam | 51.6 | 12.1 |
| ARC-AGI 2 | 18 vendor | 15 |
| AIME 2025 | 58 vendor | 78 |
| MMMLU | 75 vendor | 78 |
| MATH 500 | 68 vendor | 72 |
| Terminal-Bench 2.1 | — | 16.9 |
| MCP Atlas | 69 | — |
| BrowseComp | 85.9 | — |
| BFCL | 62 vendor | — |
| HumanEval | 79 | — |
| Independent composite scores each on its own scale — not part of the average above | ||
| LMSYS Chatbot Arena | 1435.4 ELO | 1410.2 ELO |
| AA Intelligence Index | 51.8 | — |
| AA Agentic Index | 48.4 | — |
| AA-Omniscience Index (−100–100) | -14.3 | — |
| GDPval-AA v2 | 1558.4 ELO | — |
| Providers | ||
| Available from |
DeepSeek — $0.140/$0.280/Mtok
Fireworks — $0.140/$0.280/Mtok
DeepInfra — $0.090/$0.180/Mtok
|
Google — $0.300/$2.50/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | DeepSeek V4 Flash | Gemini 2.5 Flash | Savings |
|---|---|---|---|
| 1M tokens | $0.18 | $0.85 | $0.68 (80%) |
| 10M tokens | $1.75 | $8.5 | $6.75 (79.4%) |
| 100M tokens | $17.5 | $85 | $67.5 (79.4%) |
| 1000M tokens | $175 | $850 | $675 (79.4%) |
When to pick which
distilled from the pricing and spec data above- cost dominates: $0.175/Mtok blended vs $0.850 — 79.4% less on the same 50/50 token mix
- your workload re-reads context (agents, RAG, long chats): cached input costs $0.003/Mtok — 98% off its list input price, a discount Gemini 2.5 Flash doesn't offer
- raw capability matters more than price: 77.5 vs 50.2 average benchmark score
- provider flexibility: available from 3 providers, so you can price-shop hosts or fail over
- this pairing gives Gemini 2.5 Flash no edge on price, context, cache discount, or measured performance — pick it only for qualitative fit (ecosystem, compliance, model behavior on your prompts)
Price movement
recorded changes since we began tracking each model · last checked Aug 15, 2026| Model | Changes | Latest move | Old → new $/M |
|---|---|---|---|
| DeepSeek V4 Flash | 0 | — no move | Stable since we began tracking — no change recorded |
| Gemini 2.5 Flash | 2 | price rise on Jul 4, 2026 | $0.075/$0.3 → $0.3/$2.5/Mtok |
The cheaper option here — DeepSeek V4 Flash costs $0.175/Mtok blended on DeepSeek.
Summary
DeepSeek V4 Flash by DeepSeek costs $0.140/Mtok input and $0.280/Mtok output, with a 1M-token context window. It supports text input and is available from 3 providers.
Gemini 2.5 Flash by Google costs $0.300/Mtok input and $2.50/Mtok output, with a 1M-token context window. It supports text, image, audio, video input.
On a blended cost basis, DeepSeek V4 Flash is 79.4% cheaper than Gemini 2.5 Flash.
On benchmarks, DeepSeek V4 Flash scores higher (77.5 vs 50.2) on average. In terms of value, DeepSeek V4 Flash has better performance per dollar (442.9 vs 59.1).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.