DeepSeek V4.1 Flash vs DeepSeek V4 Pro
| Spec | DeepSeek V4.1 Flash | DeepSeek V4 Pro |
|---|---|---|
| Respan Index | 52.5 (#37) | 49.2 (#45) |
| Blended price per 1M | $0.089 | $0.585 |
| Context window | 1M | 1M |
| Input / output per 1M | $0.051 / $0.203 | $0.28 / $1.5 |
| Max output | 384K | 384K |
| Input | text, vision | text |
| Released | 2026-09-10 | 2026-08-13 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | DeepSeek V4.1 Flash | DeepSeek V4 Pro |
|---|---|---|
| Coding | 43.7 DeepSeek V4.1 Flash | 29.8 DeepSeek V4 Pro |
| Agents | 58.4 DeepSeek V4.1 Flash | 59.7 DeepSeek V4 Pro |
| Reasoning | 55.5 DeepSeek V4.1 Flash | 52.8 DeepSeek V4 Pro |
| Knowledge & language | 44.5 DeepSeek V4.1 Flash | 52.6 DeepSeek V4 Pro |
| Human preference | 57.5 DeepSeek V4.1 Flash | 55.7 DeepSeek V4 Pro |
Benchmarks both report
DeepSeek V4.1 Flash leads on 15 of the 23 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | DeepSeek V4.1 Flash | DeepSeek V4 Pro |
|---|---|---|
| LMArena Elo | 1475 | 1464 |
| LMArena WebDev (LMArena) | 1659 | 1581 |
| LMArena Agent (LMArena) | 0.0078 | -0.003 |
| MirrorCode (Epoch AI) | 64.4% | 33.3% |
| APEX-Agents (Mercor) | 39.5% | 47.3% |
| FrontierMath Tier 4 v2 (Epoch AI) | 26.8% | 26.8% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 67.4% | 64.6% |
| GPQA Diamond | 90.9% | 91.7% |
| Mystery Game Puzzles (Epoch AI) | 43% | 43% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.3% | 98.6% |
| Terminal-Bench 4.0 (Vals AI) | 19.7% | 14.1% |
| Chess Puzzles (Epoch AI) | 48% | 46% |
| SimpleQA Verified | 42.9% | 52.9% |
| LiveBench Agentic Coding (LiveBench) | 77.3% | 54.9% |
| LiveBench Coding (LiveBench) | 80% | 77.2% |
| LiveBench Data Analysis (LiveBench) | 79.3% | 79.2% |
| LiveBench Instruction Following (LiveBench) | 70% | 67.7% |
| LiveBench Language (LiveBench) | 81.2% | 82.1% |
| LiveBench Mathematics (LiveBench) | 93.3% | 95.1% |
| LiveBench Reasoning (LiveBench) | 86.7% | 85.8% |
| SimpleBench (SimpleBench) | 66.7% | 50.9% |
| DeepSWE v1.1 | 74.2% | 62.7% |
| Terminal-Bench 2.1 | 90.6% | 87.9% |
FAQ
Is DeepSeek V4.1 Flash better than DeepSeek V4 Pro?
DeepSeek V4.1 Flash scores 52.5 on the Respan Index (#37), ahead of DeepSeek V4 Pro at 49.2 (#45). DeepSeek V4.1 Flash leads on 15 of the 23 benchmarks both report.
Which is cheaper, DeepSeek V4.1 Flash or DeepSeek V4 Pro?
DeepSeek V4.1 Flash lists at $0.051 input and $0.203 output per 1M tokens; DeepSeek V4 Pro at $0.28 and $1.5. DeepSeek V4.1 Flash is 6.6x cheaper per token at list price.
Which has the larger context window?
Both accept 1M tokens of context.
Compare other models
- vsDeepSeek V4.1 Flash vs DeepSeek V4 Flash
- vsGemini 3.8 Flash vs DeepSeek V4.1 Flash
- vsClaude Sonnet 5.5 vs DeepSeek V4 Pro
- vsDeepSeek V4.1 Flash vs Qwen3.8-Max
- vsDeepSeek V4.1 Flash vs MiniMax M3.1 Flash
- vsDeepSeek V4.1 Flash vs Kimi K3
- vsDeepSeek V4 Pro vs GPT-5.4
- vsDeepSeek V4 Pro vs GPT-6 Luna
- vsDeepSeek V4 Pro vs Claude Opus 4.7