Compare / DeepSeek vs MiniMax
DeepSeek vs MiniMax: which AI lab is ahead?
Each lab's best model on sale goes head to head on every kind of work we measure, from our own benchmarks and licensed sources. 16 models in all; a lab is only as good as the model you would actually pick from it.
Kinds of work led by each lab's best model there, of the 10 both are measured on.
Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331). Models on sale means listed on OpenRouter today.
DeepSeek or MiniMax? The short answer
DeepSeek is ahead: its best models beat MiniMax's in 7 of the 10 kinds of work both are measured on; MiniMax leads in 3.
- Pick DeepSeek for user surveys, marketing planning, agents and tool use, professional work, writing, sticking to the facts and speed under pressure.
- Pick MiniMax for product listings, decks from an analysis and reasoning and knowledge.
- On our overall leaderboard, DeepSeek-V4.1-Flash is #21 (49 of 100) and MiniMax M3.1 Flash is #25 (40).
- Prices start at $0.15 per million tokens for DeepSeek (DeepSeek-V4-Flash (0423)) and $0.26 for MiniMax (MiniMax M3.1 Flash).
Head to head
Each line names the model behind the number. Prices are blended, three input tokens to one output, at the developer's list price where we have read it; see every model's price. Intelligence Index from Artificial Analysis; overall score from our leaderboard.
Which lab is better for what
For each kind of work, each lab's best model there and its rank among every model on sale. Further right is better; the better rank is in green.
Every model, by intelligence and price
Each dot is a model on sale with an Intelligence Index from Artificial Analysis. Up is smarter; left is cheaper. Hover or tap a dot for its name.
DeepSeekMiniMax
On our overall leaderboard
The 4 of their models we rank overall: one score out of 100 across everything we publish (how the score works). Only models with enough of our own results are ranked.
- #21DeepSeek-V4.1-Flash49
- #25MiniMax M3.1 Flash40
- #29DeepSeek-V4.1-Flash-101737
- #34DeepSeek-V4-Pro (0423)27
Benchmark by benchmark
The headline result of every benchmark both labs have models on: each lab's best model and its rank among the models on sale. Green marks the better rank.
| Measure | DeepSeek | MiniMax |
|---|---|---|
| Agents and tool use | ||
| OpenHands Index · Average% resolved, higher is better | 45.7%DeepSeek-V3.2 · 20 of 31 | 57.2%MiniMax-M3 · 13 of 31 |
| Artificial Analysis · Terminal-Bench 4.0% of tasks, higher is better | 33.8%DeepSeek-V4.1-Flash-1017 · 18 of 107 | 25.8%MiniMax M3.1 Flash · 23 of 107 |
| Artificial Analysis · Terminal-Bench 2.1% of tasks, higher is better | 78.7%DeepSeek-V4-Pro (0813) · 24 of 107 | 65.2%MiniMax-M3 · 38 of 107 |
| Artificial Analysis · τ-bench banking% of tasks, higher is better | 39.6%DeepSeek-V4-Pro (0813) · 17 of 104 | 15.3%MiniMax-M3 · 54 of 104 |
| Vending-Bench 2 · Money after a yearUS dollars, higher is better | $3,285DeepSeek-V4-Pro (0423) · 36 of 53 | $2,158MiniMax-M3 · 38 of 53 |
| Decks from an analysis | ||
| DeckBench · Deck ratingrating, higher is better | 1,046DeepSeek-V4.1-Flash-1017 · 22 of 35 | 1,266MiniMax M3.1 Flash · 10 of 35 |
| Marketing planning | ||
| ROASBench · Overall scorescore out of 100, higher is better | 41.4DeepSeek-V4.1-Flash-1017 · 23 of 35 | 40.3MiniMax M3.1 Flash · 24 of 35 |
| Product listings | ||
| CatalogBench · Reliably publish-ready% of products, higher is better | 41.1%DeepSeek-V4.1-Flash · 20 of 35 | 50.0%MiniMax M3.1 Flash · 17 of 35 |
| CatalogBench · Reliably publish-ready · sales brief% of products, higher is better | 23.2%DeepSeek-V4.1-Flash · 14 of 35 | 37.5%MiniMax M3.1 Flash · 10 of 35 |
| Professional work | ||
| APEX-Agents · Tasks passed% of tasks, higher is better | 47.3%DeepSeek-V4-Pro (0813) · 24 of 37 | 37.7%MiniMax-M3 · 30 of 37 |
| Reasoning and knowledge | ||
| Artificial Analysis · Artificial Analysis Intelligence Indexindex score, higher is better | 45.1DeepSeek-V4.1-Flash-1017 · 16 of 185 | 42.4MiniMax M3.1 Flash · 21 of 185 |
| Artificial Analysis · Humanity's Last Exam% of questions, higher is better | 43.3%DeepSeek-V4.1-Flash-1017 · 29 of 186 | 48.0%MiniMax M3.1 Flash · 15 of 186 |
| Artificial Analysis · GPQA Diamond% of questions, higher is better | 92.8%DeepSeek-V4-Pro (0813) · 15 of 168 | 92.9%MiniMax-M3 · 14 of 168 |
| Speed under pressure | ||
| BulletBench · Ladder Elo · Bullet 60sladder Elo, higher is better | 147DeepSeek-V4.1-Flash-1017 · 28 of 36 | 115MiniMax M3.1 Flash · 32 of 36 |
| Sticking to the facts | ||
| Vectara Hallucination Leaderboard · Hallucination rate · Vectara% of summaries, lower is better | 5.3%DeepSeek-V3.2-Exp · 9 of 71 | 9.1%MiniMax-M2.5 · 33 of 71 |
| Arena (formerly LMArena) · Overall · Text factualityArena rating, higher is better | 1,467DeepSeek-V4.1-Flash · 24 of 109 | 1,434MiniMax-M3 · 70 of 109 |
| User surveys | ||
| SurveyBench · Analysis ratingrating, higher is better | 1,119DeepSeek-V4.1-Flash-1017 · 12 of 26 | 428MiniMax M3.1 Flash · 26 of 26 |
| Writing | ||
| Arena (formerly LMArena) · Overall · TextArena rating, higher is better | 1,474DeepSeek-V4.1-Flash · 27 of 153 | 1,440MiniMax-M3 · 64 of 153 |
| UGI Leaderboard · Writing score · UGIscore out of 100, higher is better | 68.4DeepSeek-V4-Pro (0423) · 14 of 124 | 41.7MiniMax-M2.5 · 76 of 124 |
Quick answers
Is DeepSeek or MiniMax better?
DeepSeek is ahead: its best models beat MiniMax's in 7 of the 10 kinds of work both are measured on; MiniMax leads in 3.
What is DeepSeek's best model?
DeepSeek-V4.1-Flash: #21 on the Spring Prompt overall leaderboard and 39.5 on the Artificial Analysis Intelligence Index.
What is MiniMax's best model?
MiniMax M3.1 Flash: #25 on the Spring Prompt overall leaderboard and 42.4 on the Artificial Analysis Intelligence Index.
Which is better for agents and tool use, DeepSeek or MiniMax?
DeepSeek: its best model there, DeepSeek-V4.1-Flash-1017, ranks 24 of 159 models on sale, against 33 for MiniMax's MiniMax M3.1 Flash.
Which is better for reasoning and knowledge, DeepSeek or MiniMax?
MiniMax: its best model there, MiniMax M3.1 Flash, ranks 19 of 187 models on sale, against 24 for DeepSeek's DeepSeek-V4.1-Flash-1017.
Which is better for writing, DeepSeek or MiniMax?
DeepSeek: its best model there, DeepSeek-V4.1-Flash, ranks 29 of 169 models on sale, against 67 for MiniMax's MiniMax-M3.
Which is cheaper, DeepSeek or MiniMax?
DeepSeek's cheapest model with results, DeepSeek-V4-Flash (0423), costs $0.15 per million tokens (three input to one output), against $0.26 for MiniMax M3.1 Flash from MiniMax.
Other labs compared
- DeepSeek vs OpenAIBest model against best model
- Anthropic vs DeepSeekBest model against best model
- DeepSeek vs GoogleBest model against best model
- DeepSeek vs xAIBest model against best model
- DeepSeek vs MetaBest model against best model
- Alibaba vs DeepSeekBest model against best model
- DeepSeek vs Mistral AIBest model against best model
- DeepSeek vs Z.aiBest model against best model
Which model is best for your product?
Benchmarks aren't your task. Run models from DeepSeek, MiniMax and the rest on your own prompt, and see the cheapest one that passes.