Models / DeepSeek-V4.1-Flash-1017
DeepSeek
DeepSeek-V4.1-Flash-1017
DeepSeek-V4.1-Flash-1017 is best for reasoning and knowledge, and agents and tool use; a strong pick at a modest price. The cheapest model of similar intelligence is Claude Haiku 5.5, at $0.20 per million tokens against $0.52.
Results 20 Oct 2026 · 92 results · 6 sources
- Price per million tokens
- $0.30 in · $1.20 out
DeepSeek's list price, 13 Oct 2026 · compare prices$0.01 cached input - Speed
- 212 tokens a second; on its test prompt the answer starts after 10.5 s, thinking time included
Artificial Analysis, on its usual API, at max reasoning - Intelligence Index
- 45.1
Artificial Analysis, at max reasoning - Spring Prompt overall
- 16th of 27 score 44 of 100 · how it works
- Developer
- DeepSeek, based in China
Weights published - Released
- 17 Oct 2026
1,048,576 tokens of context - Pricing and availability
- Half price off-peak ($0.15 / $0.60)
Where it stands among the models you could choose
Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.
| Among | Intelligence Index | Spring Prompt overall | Price, blended | Speed |
|---|---|---|---|---|
| Models available today | 16th of 202Claude Opus 5.5 | 16th of 27Claude Opus 5.5 | $0.52 blended, 37th percentilecheapest: Mistral Nemo | 12th of 64Gemini 3.5 Flash-Lite |
| Models priced $0.25–1 per million tokens | 2nd of 70MiMo-V2.6-Pro | 3rd of 5DeepSeek-V4.1-Flash | sets this group | 6th of 22Gemini 3.5 Flash-Lite |
| Models with similar intelligence | sets this group | 6th of 9GPT-6 Sol | $0.52 blended, 21st percentilecheapest: Claude Haiku 5.5 | 3rd of 11Muse Spark 1.3 |
| Models writing 200 or more tokens a second | 2nd of 13Muse Spark 1.3 | 3rd of 5Muse Spark 1.3 | $0.52 blended, 58th percentilecheapest: gpt-oss-20b | sets this group |
| Models from developers based in China | 3rd of 88MiMo-V2.6-Pro | 5th of 8DeepSeek-V4.1-Flash | $0.52 blended, 42nd percentilecheapest: Qwen3.5-9B | 2nd of 24DeepSeek-V4.1-Flash |
| Open-weights modelsincluding 2 with weights announced but not yet published | 2nd of 112MiMo-V2.6-Pro | 4th of 7DeepSeek-V4.1-Flash | $0.52 blended, 54th percentilecheapest: Mistral Nemo | 7th of 42DeepSeek-V4.1-Flash |
The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.
Measured cost per task
API cost recorded in our benchmark runs, excluding judging.
| Task | Benchmark | Cost per task |
|---|---|---|
| Product listing | CatalogBench | ≈ $0.002 |
| Deck | DeckBench | ≈ $0.026 |
| Survey plan and analysis | SurveyBench | ≈ $0.037 |
| Game of bullet chess | BulletBench | ≈ $0.005 |
How good is it, and for what?
Each use case ranks the available models its benchmarks measured, then those priced $0.25–1 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.
Product listingsTurning a sparse product feed and photos into listings that can go live
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| CatalogBenchCatalogBench: reliably publish-ready | 21.4% | 19th of 24 | GPT-6 Astra 75.0% |
| CatalogBenchCatalogBench (sales brief): reliably publish-ready | 0.0% | 21st of 24 | GPT-6 Astra 69.6% |
Decks from an analysisTurning a finished analysis into a deck you could present as it is
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| DeckBenchDeckBench: deck rating | 1,145 | 12th of 24 | GPT-6 Astra 1,577 |
User surveysPlanning a user survey and reading its results without being misled
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| SurveyBenchSurveyBench: analysis rating | 1,221 | 5th of 13 | Claude Opus 5.5 1,499 |
Marketing planningPlanning a year of ad spend without overspending
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| ROASBenchROASBench: overall score | 41.4 | 11th of 24 | GPT-6 Astra 56.6 |
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported | 33.8% | 16th of 109 | Claude Sonnet 5.5 63.6% |
- tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
- Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
- OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
- Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
- Vending-Bench 2: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects
- APEX-Agents: pending. Its latest results are dated 1 Oct 2026, before this model was released.
- GDP.pdf: pending. Its latest results are dated 1 Oct 2026, before this model was released.
- Remote Labor Index: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Reasoning and knowledgeHard questions across science, maths and general knowledge
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported | 45.1 | 16th of 202 | Claude Opus 5.5 57.6 |
| Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported | 43.3% | 26th of 201 | Claude Opus 5.5 61.4% |
WritingWhat people prefer in blind comparisons, and judged writing quality
- Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
- UGI Leaderboard: pending. Its latest results are dated 2 Oct 2026, before this model was released.
Sticking to the factsSummarising without inventing things, and factual answers
- Vectara Hallucination Leaderboard: pending. Its latest results are dated 22 Sep 2026, before this model was released.
- Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
- SimpleQA Verified (Epoch AI): pending. Its latest results are dated 29 Sep 2026, before this model was released.
Speed under pressureGood decisions against a real clock (fast chess)
| Benchmark | DeepSeek-V4.1-Flash-1017 | Rank among available | Best available |
|---|---|---|---|
| BulletBenchBulletBench Bullet 60s: ladder Elo | 147 | 20th of 25 | Gemini 3.8 Flash 1,068 |
Nearest alternatives
- Cheaper at similar intelligenceClaude Haiku 5.5$0.20 blended · Intelligence Index 43.4Compare with DeepSeek-V4.1-Flash-1017 →
Against DeepSeek-V4.1-Flash-1017 at $0.52 blended, Intelligence Index 45.1. "Similar intelligence" means within 4 points or better.
Against the alternatives
The models you would most likely weigh it against: the leaders of the groups above.
Scroll sideways to see every alternative.
| DeepSeek-V4.1-Flash-1017 | MiMo-V2.6-Pro most intelligent at $0.25–1 per million tokens | Claude Haiku 5.5 cheapest with similar intelligence | Muse Spark 1.3 fastest with similar intelligence | Claude Opus 5.5 most intelligent available today | |
|---|---|---|---|---|---|
| Price per million tokens | $0.52 | $0.54 | $0.20 | $2.00 | $8.00 |
| Tokens a second | 212 | 43 | 240 | 254 | 96 |
| Intelligence Index | 45.1 | 46.3 | 43.4 | 48.1 | 57.6 |
| Spring Prompt overall | 44 | – | 43 | 62 | 85 |
| Rank among available models, by use case | |||||
| Product listings | 21st of 24 | – | 14th | 12th | 9th |
| Decks from an analysis | 12th of 24 | – | 11th | 16th | 4th |
| User surveys | 5th of 13 | – | 4th | – | 1st |
| Marketing planning | 11th of 24 | – | 22nd | 16th | 5th |
| Agents and tool use | 22nd of 168 | 19th | 25th | 11th | 2nd |
| Professional work | pending | – | – | 8th | 3rd |
| Reasoning and knowledge | 22nd of 202 | 12th | 23rd | 10th | 1st |
| Writing | pending | 23rd | – | 7th | 5th |
| Sticking to the facts | pending | 9th | – | 12th | 3rd |
| Speed under pressure | 21st of 27 | – | 17th | 25th | – |
| Side by side → | Side by side → | Side by side → | |||
Shaded figures are better than DeepSeek-V4.1-Flash-1017's. A dash means no result.
Research and examples
- DeepSeek-V4.1-Flash-1017 review: better at analysis, much worse at sticking to the product data
- Mercury 3 first look: very fast with thinking off, too slow with it on
- MiniMax M3.1 Flash first look: careful with product data, unreliable on long jobs
In the playground: our pick in 1 of 29 tests, 89% of checks passed →
- Humpback whale breach in three.js (our pick)
- Throw a bowl on a potter's wheel from a lump of wet clay (4 of 4 checks)
- Paint a lemon on a blue plate as a thick oil painting on canvas (4 of 4 checks)
Where it has been measured
2nd most intelligent of the 112 open-weights models (including 2 with weights announced but not yet published), and 2nd fastest of the 24 models from developers based in China. 2nd most intelligent of the 70 models priced $0.25–1 per million tokens. Among models available today, in the top quarter (by percentile) for reasoning and knowledge, and agents and tool use; in the bottom quarter for product listings and speed under pressure. Results for professional work, writing and sticking to the facts are pending: those sources last updated before this model was released.
- BulletBench5 results
- CatalogBench32 results
- DeckBench18 results
- ROASBench9 results
- SurveyBench21 results
- Artificial Analysis7 results
- APEX-AgentsPending: latest results 1 Oct 2026
- Arena (formerly LMArena)Pending: latest results 2 Oct 2026
- Berkeley Function Calling Leaderboard (BFCL) V4Pending: latest results 16 Dec 2025
- GDP.pdfPending: latest results 1 Oct 2026
- Microsoft STATE-BenchPending: latest results 1 Jul 2026
- OpenHands IndexPending: latest results 30 Jun 2026
- Remote Labor IndexPending: latest results 1 Oct 2026
- SimpleQA Verified (Epoch AI)Pending: latest results 29 Sep 2026
- tau2-benchPending: latest results 3 Aug 2026
- UGI LeaderboardPending: latest results 2 Oct 2026
- Vectara Hallucination LeaderboardPending: latest results 22 Sep 2026
- Vending-Bench 2Pending: latest results 1 Oct 2026
Every result
92 published results from 6 sources, each in the source's own units, with the configuration that produced it.
BulletBench · measured by Spring Prompt · 5 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| BulletBench Bullet 60s: cost per gameUS dollars, lower is better | $0.0047 | 14 of 40 | DeepSeek-V4.1-Flash-1017 (low reasoning) low reasoning |
19 Oct 2026 |
| BulletBench Bullet 60s: games lost on time% of games, lower is better | 87.5% | 34 of 40 | DeepSeek-V4.1-Flash-1017 (low reasoning) low reasoning |
19 Oct 2026 |
| BulletBench Bullet 60s: invalid moves% of moves, lower is better | 0.0% | 1 of 40 | DeepSeek-V4.1-Flash-1017 (low reasoning) low reasoning |
19 Oct 2026 |
| BulletBench Bullet 60s: ladder Eloladder Elo, higher is better | 147 0–443 |
32 of 40 | DeepSeek-V4.1-Flash-1017 (low reasoning) low reasoning |
19 Oct 2026 |
| BulletBench Bullet 60s: median move timemilliseconds, lower is better | 3.3 s | 37 of 40 | DeepSeek-V4.1-Flash-1017 (low reasoning) low reasoning |
19 Oct 2026 |
Not shown: Not your cost: prices are those charged on the run date.
CatalogBench · measured by Spring Prompt · 32 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| CatalogBench (sales brief): channel rules broken% of products, lower is better | 2.4% | 22 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): claims to check% of products, lower is better | 3.6% | 18 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): content quality% of checks, higher is better | 94.8% | 19 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): failed outputs% of products, lower is better | 0.0% | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): missing UK information% of products, lower is better | 0.0% | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): not findable% of products, lower is better | 2.4% | 7 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): publish-ready listings% of products, higher is better | 8.3% | 21 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): reliably publish-ready% of products, higher is better | 0.0% | 21 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): unsupported claims% of products, lower is better | 85.7% | 23 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): unsupported claimsclaims per product, lower is better | 4.01 | 22 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): wrong attributes% of products, lower is better | 22.0% | 21 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench (sales brief): wrong category or variant% of products, lower is better | 1.2% | 3 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: channel compliance% of products, higher is better | 100.0% | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: channel rules broken% of products, lower is better | 1.2% | 19 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: claims to check% of products, lower is better | 1.2% | 19 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: conflicts caught% of conflicts, higher is better | 100.0% | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: content quality% of checks, higher is better | 95.3% | 20 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: cost per productUS dollars, lower is better | $0.0015 | 4 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: decision accuracy% of decisions, higher is better | 94.9% | 18 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: failed outputs% of products, lower is better | 0.0% | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: field accuracy% of missing fields, higher is better | 93.7% | 10 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: invented values% of filled values, lower is better | 1.7% | 13 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: missing UK information% of products, lower is better | 0.0% | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: not findable% of products, lower is better | 3.6% | 5 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: output tokens per producttokens, lower is better | 1,226 | 1 of 2 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: publish-ready listings% of products, higher is better | 40.5% | 19 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: reliably publish-ready% of products, higher is better | 21.4% | 19 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: time per productseconds, lower is better | 6.70 | 3 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: unsupported claims% of products, lower is better | 44.6% | 20 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: unsupported claimsclaims per product, lower is better | 1.70 | 21 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: wrong attributes% of products, lower is better | 17.9% | 13 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| CatalogBench: wrong category or variant% of products, lower is better | 0.6% | 2 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
DeckBench · measured by Spring Prompt · 18 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| DeckBench: accurate decks% of tasks, higher is better | 33.3% | 14 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: caveat dropped% of tasks, lower is better | 16.7% | 12 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: clean layout% of tasks, higher is better | 16.7% | 20 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: cost per deckUS dollars, lower is better | $0.0255 | 7 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: deck ratingrating, higher is better | 1,145 1,020–1,233 |
12 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: design quality% of the maximum, higher is better | 68.5% | 9 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: draft figure quoted% of tasks, lower is better | 50.0% | 19 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: findings missing% of tasks, lower is better | 0.0% | 1 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: head-to-head win rate% of comparisons, higher is better | 53.3% | 14 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: layout defects% of tasks, lower is better | 83.3% | 20 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: misleading metric used% of tasks, lower is better | 0.0% | 1 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: output tokens per decktokens, lower is better | 21,019 | 1 of 3 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: presentable decks% of tasks, higher is better | 0.0% | 12 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: recommendation late or wrong% of tasks, lower is better | 16.7% | 9 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: slides needing work% of tasks, lower is better | 66.7% | 16 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: time per deckseconds, lower is better | 73.0 | 7 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: unsupported claims% of tasks, lower is better | 16.7% | 11 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| DeckBench: unsupported numbers% of tasks, lower is better | 0.0% | 1 of 24 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.
ROASBench · measured by Spring Prompt · 9 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| ROASBench: audience scorescore out of 100, higher is better | 65.3 | 9 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: business scorescore out of 100, higher is better | 38.3 | 11 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: consistency scorescore out of 100, higher is better | 20.2 | 15 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: contribution profitsimulated US dollars, higher is better | $585,572 | 10 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: cost of a runUS dollars, lower is better | $0.0799 | 8 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: months over budgetmonths of 12, lower is better | 0 | 1 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: overall scorescore out of 100, higher is better | 41.4 | 11 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: planning scorescore out of 100, higher is better | 54.3 | 21 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| ROASBench: return on ad spendprofit per $1 spent, higher is better | $0.44 | 12 of 25 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.
SurveyBench · measured by Spring Prompt · 21 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| SurveyBench: analysis ratingrating, higher is better | 1,221 1,116–1,327 |
5 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: analysis scorepoints of 100, higher is better | 90.2 | 4 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: cost per taskUS dollars, lower is better | $0.0373 | 4 of 12 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: head-to-head win rate% of comparisons, higher is better | 60.2% | 5 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: leading or double-barrelled questions% of tasks, lower is better | 11.1% | 3 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: mistaken next steps% of tasks, lower is better | 0.0% | 1 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: numbers right% of answers, higher is better | 100.0% | 1 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: objectives unanswerable% of tasks, lower is better | 11.1% | 7 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: output tokens per tasktokens, lower is better | 30,893 | 1 of 2 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: precision ignored% of tasks, lower is better | 0.0% | 1 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: research scorepoints of 100, higher is better | 90.9 | 4 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: sound research% of tasks, higher is better | 33.3% | 3 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: stakeholder's view presumed% of tasks, lower is better | 0.0% | 1 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: survey plan scorepoints of 100, higher is better | 92.5 | 4 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: survey structure% of tasks, lower is better | 33.3% | 10 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: time per taskseconds, lower is better | 134 | 6 of 12 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: traps handled% of traps, higher is better | 91.3% | 5 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: traps missed% of tasks, lower is better | 44.4% | 5 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: unsupported findings% of tasks, lower is better | 44.4% | 4 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: wrong numbers% of tasks, lower is better | 0.0% | 1 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
| SurveyBench: wrong recommendation% of tasks, lower is better | 11.1% | 3 of 13 | DeepSeek-V4.1-Flash-1017 provider default reasoning |
19 Oct 2026 |
Not shown: Not your customers or your data: invented organisations with simulated respondents, nine public tasks, one run per model per task at the provider's default reasoning setting.
Artificial Analysis · reported by Artificial Analysis · 7 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better | 45.1 | 35 of 413 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
| Artificial Analysis: Humanity's Last Exam% of questions, higher is better | 43.3% | 55 of 411 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
| Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better | 82.7% | 30 of 399 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
| Artificial Analysis: Output speedtokens per second, higher is better | 212 | 20 of 137 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
| Artificial Analysis: SciCode% of problems, higher is better | 55.9% | 40 of 197 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better | 33.8% | 36 of 193 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
| Artificial Analysis: Time to first answer tokenseconds, lower is better | 10.5 | 61 of 137 | DeepSeek-V4.1-Flash-1017 (max reasoning) | 20 Oct 2026 |
Not shown: Not business work, and a blend: read the parts for any one task.