Quoted word for word from the vendor's own launch post or model card. Claims, not measurements.
No vendor claims recorded yet. They come from launch posts and model cards the Drops covered.
Independent results
Measured by independent sources (Epoch AI, LMArena, OpenRouter's benchmark runs, τ²-bench, Vectara), not by the vendor. "Listed as" is the source's own name for the model, so you can check our match.
Independent benchmark results for this model, each with its source and date
Board
Result
Rank
Epoch AI · MirrorCodeListed as: deepseek-v4.1-flash_maxSource: Epoch AI (CC BY), as of Nov 20, 2026. Source
64.4%Accuracy
12
Epoch AI · Chess PuzzlesListed as: deepseek-v4.1-flash_maxSource: Epoch AI (CC BY), as of Nov 20, 2026. Source
48.0%Accuracy
23
Epoch AI · Mystery Game PuzzlesListed as: deepseek-v4.1-flash_maxSource: Epoch AI (CC BY), as of Nov 20, 2026. Source
43.0%Accuracy
21
Epoch AI · Furniture AssemblyListed as: deepseek-v4.1-flash_maxSource: Epoch AI (CC BY), as of Nov 20, 2026. Source
34.2%Accuracy
27
LMArena · WebDevListed as: DeepSeek-V4.1-Flash-20261017 (Max)LMArena (CC BY 4.0), as of Nov 18, 2026. Source
1,659Arena score
22
OpenRouter · GPQA DiamondListed as: DeepSeek V4.1 Flashvia NextBit, the most accurate providerSource: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source
90.9%Accuracy
29
OpenRouter · GPQA Diamond: cost per taskListed as: DeepSeek V4.1 Flashvia NextBit, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source
$0.026Cost per task
17
OpenRouter · GPQA Diamond: time per taskListed as: DeepSeek V4.1 Flashvia NextBit, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source
178 sTime per task
36
OpenRouter · τ²-bench airlineListed as: DeepSeek V4.1 Flashvia Crusoe, the most accurate providerSource: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source
76.3%Accuracy
30
OpenRouter · τ²-bench airline: cost per taskListed as: DeepSeek V4.1 Flashvia Crusoe, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source
$0.033Cost per task
13
OpenRouter · τ²-bench airline: time per taskListed as: DeepSeek V4.1 Flashvia Crusoe, the most accurate providerLower is better on this board.Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0, as of Nov 20, 2026. Source
145 sTime per task
20
Vectara · Hallucination rate in summariesListed as: deepseek-ai/DeepSeek-V4.1-FlashLower is better on this board.Vectara Hallucination Leaderboard (HHEM-2.3), Apache-2.0, as of Oct 22, 2026. Source
4.2%Percent
6
Price history
OpenRouter's list price per 1M tokens in each of our snapshots.
$3$0
Oct 5, 2026Nov 20, 2026
Output now $1.2
Input now $0.3
Changed 44 times across our 93 snapshots.
Show the numbers
This model's OpenRouter price in each of our snapshots
Epoch AI's new runs give deepseek-v4.1-flash_max 0.898 on GPQA diamond, 0.268 on FrontierMath Tier 4 and 0.983 on OTIS Mock AIME. gpt-6-astra_max and gpt-6.1-sol_max lead Tiers 1–3 at 0.937.
Why it matters: Independent scores show how far DeepSeek's Flash model trails the frontier on hard math, useful before routing such work to it.