DeepSeek V4.1 Flash 1017 vs Inkling 2 Small
Full profiles: DeepSeek V4.1 Flash 1017 (DeepSeek) and Inkling 2 Small (Thinking Machines): every benchmark, price and rank.
Updated · Data from 14 sources
As of 16 Nov 2026, DeepSeek V4.1 Flash and Inkling 2 Small are neck and neck on quality (57.7 vs 57.9 on the BenchLeader Index, inside the ±4.9 margin), and they cost about the same at $0.525 vs $0.475 per 1M tokens blended. Choose DeepSeek V4.1 Flash for reasoning (56 vs 53) and coding (63 vs 60), where it also streams ~1.5× faster; choose Inkling 2 Small for knowledge (54 vs 48), agents & tools (61 vs 58), and multimodal (64 vs 63).
Key facts (as of 16 Nov 2026)
- DeepSeek V4.1 Flash scores 57.7 on the BenchLeader Index and Inkling 2 Small scores 57.9 (BenchLeader, 16 Nov 2026).
- DeepSeek V4.1 Flash costs $0.525 per 1M tokens blended (list price, 16 Nov 2026).
- Inkling 2 Small costs $0.475 per 1M tokens blended (list price, 16 Nov 2026).
- DeepSeek V4.1 Flash streams ~1.5× faster than Inkling 2 Small (Artificial Analysis, 7-day median, 16 Nov 2026).
- Of the 35 benchmarks both models have, Inkling 2 Small leads 19 (BenchLeader, 16 Nov 2026).
- On Arena WebDev, DeepSeek V4.1 Flash 1017 is rated 1664 and Inkling 2 Small 1647 (LMArena, as published by 16 Nov 2026).
Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | DeepSeek V4.1 Flash 1017 | 63 vs 60 category score; Arena WebDev 1664 vs 1647 | Medium |
| Agents & tool use | Inkling 2 Small | 61 vs 58 category score; Arena Agent 2.5 vs 2 | Medium |
| Reasoning | DeepSeek V4.1 Flash 1017 | 56 vs 53 category score; HLE-Diamond (AA) 25.3% vs 28.7% | Medium |
| Knowledge | Inkling 2 Small | 54 vs 48 category score; AA-Omniscience v1.1 16.1 vs 0.9 | Medium |
| Long documents | Either | 66 vs 66 category score; AA-LCR v1.1 82.7% vs 83.3% | Either |
| Images and other media | Inkling 2 Small | 64 vs 63 category score; MMMU-Pro 81.6% vs 80.6% | Medium |
| Chat quality (human preference) | – | Only Inkling 2 Small has an Arena rating (59 category score) | Not comparable |
| Lowest cost | Either | $0.525 vs $0.475 per 1M blended; agentic coding session $0.023 vs $0.022 | – |
| Fastest first answer | Either | 39.7 s (DeepSeek V4.1 Flash 1017 (max)) vs 49.0 s (Inkling 2 Small) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- DeepSeek V4.1 Flash 1017 (max) and Inkling 2 Small are neck and neck on quality (57.7 vs 57.9, within the ±4.9 margin).
- DeepSeek V4.1 Flash 1017 (max) is stronger in coding, reasoning.
- Inkling 2 Small is stronger in agents & tools, knowledge, long context, multimodal.
- They cost about the same ($0.475 per 1M blended).
- DeepSeek V4.1 Flash 1017 (max) streams 1.7× faster (174 vs 100 tokens per second).
At a glance
| Spec | DeepSeek V4.1 Flash 1017 | Inkling 2 Small |
|---|---|---|
| Developer | DeepSeek | Thinking Machines |
| Released | 17 Oct 2026 | 5 Nov 2026 |
| Context window (tokens) | 1,048,576 | 1,048,576 |
| Compared at | DeepSeek V4.1 Flash 1017 (max) | Inkling 2 Small |
| Open weights | Yes | No |
| Price in / out / cached input per 1M | $0.300 / $1.20 / $0.0060 | $0.300 / $1.00 / $0.030 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | DeepSeek V4.1 Flash 1017 (max) | Inkling 2 Small | Leader |
|---|---|---|---|
| Reasoning | 55.7 | 52.8 | DeepSeek V4.1 Flash 1017 |
| Coding | 63.0 | 60.3 | DeepSeek V4.1 Flash 1017 |
| Agents & tools | 57.6 | 61.1 | Inkling 2 Small |
| Maths | 60.9 | – | not comparable |
| Knowledge | 47.7 | 53.5 | Inkling 2 Small |
| Human preference | – | 58.6 | not comparable |
| Multimodal | 63.0 | 64.0 | Inkling 2 Small |
| Long context | 66.0 | 66.4 | level |
| Composite | 55.5 | – | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Maths, Instruction following, Human preference.
Where sources disagree
Across the 35 benchmarks both have, DeepSeek V4.1 Flash 1017 leads 16, Inkling 2 Small leads 19. Every overall index that rates both puts DeepSeek V4.1 Flash 1017 ahead.
| Overall index | DeepSeek V4.1 Flash 1017 (max) | Inkling 2 Small | Leader |
|---|---|---|---|
| BenchLeader Index | 57.7 | 57.9 | neck and neck |
| AA Intelligence Index v5.0.1 | 32.6 | 30.6 | DeepSeek V4.1 Flash 1017 |
| Vals Index | 61.3 | 56.1 | DeepSeek V4.1 Flash 1017 |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | DeepSeek V4.1 Flash 1017 (max) | Inkling 2 Small |
|---|---|---|
| Blended price per 1M tokens | $0.525 | $0.475 |
| Input / output per 1M | $0.300 / $1.20 | $0.300 / $1.00 |
| Cached input per 1M | $0.0060 | $0.030 |
| Price per task: chat reply | $0.0005 | $0.0004 |
| Price per task: summarize a 30-page report | $0.0043 | $0.0042 |
| Price per task: code edit | $0.0036 | $0.0033 |
| Price per task: agentic coding session | $0.023 | $0.022 |
| Price per task: structured extraction | $0.0008 | $0.0008 |
| Agentic coding session, input mostly cached | $0.0096 | $0.0099 |
| Cost per benchmark task (AA Index) | $0.602 | $1.18 |
| Output tokens per task (AA Index) | 139k | 150k |
| Of which reasoning | 91k | 88k |
| Time per task (AA Index) | 13 min | 25 min |
| Output speed | 174 tok/s | 100 tok/s |
| Time to first answer | 39.7 s | 49.0 s |
| Time to a complete answer, last measured 13 Nov 2026 | 91.2 s | 163.6 s |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $0.602 on DeepSeek V4.1 Flash 1017 (max) and $1.18 on Inkling 2 Small, and Inkling 2 Small wrote 1.1× as many output tokens for the same work.
Each model is compared at its best-ranked reasoning-effort setting. DeepSeek V4.1 Flash 1017 (max) scores 57.7 and takes 40 s to its first answer token; DeepSeek V4.1 Flash 1017 (high) scores 51.2 in 20 s, DeepSeek V4.1 Flash 1017 (no reasoning) scores 44.9 in 1 s. Inkling 2 Small scores 57.9 and takes 49 s to its first answer token. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
120 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | DeepSeek V4.1 Flash 1017 (max) | Inkling 2 Small |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 57.7 | 57.9 |
| Agents & tools score | BenchLeader | 57.6 | 61.1 |
| Coding score | BenchLeader | 63.0 | 60.3 |
| Composite score | BenchLeader | 55.5 | – |
| Knowledge score | BenchLeader | 47.7 | 53.5 |
| Long context score | BenchLeader | 66.0 | 66.4 |
| Maths score | BenchLeader | 60.9 | – |
| Multimodal score | BenchLeader | 63.0 | 64.0 |
| Reasoning score | BenchLeader | 55.7 | 52.8 |
| Human preference score | BenchLeader | – | 58.6 |
| Blended price $/M | BenchLeader | $0.525 | $0.475 |
| Output speed | BenchLeader | 174 tok/s | 100 tok/s |
| Time to first answer | BenchLeader | 39.7 s | 49.0 s |
| Context window | BenchLeader | 1,048,576 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 93.2% | – |
| Arena Hard Prompts | Arena (LMArena) | – | 1438 |
| LiveBench 2.0 Reasoning | LiveBench | 89.1% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 43.3% | 44.1% |
| HLE-Diamond (AA) | Artificial Analysis | 25.3% | 28.7% |
| ARC-AGI-1 | ARC Prize | 93.5% | 96.5%† |
| ARC-AGI-2 | ARC Prize | 78.1% | 81.7%† |
| ARC-AGI-3 | ARC Prize | 5.8% | – |
| CritPt | Artificial Analysis | 24.0% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | 25.0% | 35.0% |
| MysteryMechanism | Vals AI | – | 25.2% |
| HLE-Diamond (Scale) | Scale AI SEAL | 20.5%† | – |
| Arena Expert | Arena (LMArena) | – | 1488 |
| Chess Puzzles | Epoch AI Benchmarking Hub | 46.0% | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 49.0% | – |
| Coding | |||
| SciCode | SciCode | 55.9% | 52.3% |
| Arena Coding | Arena (LMArena) | – | 1485 |
| Arena WebDev | Arena (LMArena) | 1664 | 1647 |
| LiveBench 2.0 Coding | LiveBench | 80.4% | – |
| Code Migration | Vals AI | 44.2% | – |
| Code Migration v1.1 | Vals AI | 67.5% | – |
| ProgramBench | Vals AI | – | 0.5% |
| SWE-bench Multimodal | Vals AI | – | 43.1% |
| Vibe Code Bench v1.1 | Vals AI | 87.7% | 81.5% |
| Vibe Code Bench 1-100 v1.1 | Vals AI | 26.1% | – |
Show all 120 benchmarks
| Metric | Publisher | DeepSeek V4.1 Flash 1017 (max) | Inkling 2 Small |
|---|---|---|---|
| Coding | |||
| SWE Atlas v1.1 | Scale AI SEAL | – | 61.9% |
| SWE Atlas v1.1: Codebase QnA | Scale AI SEAL | – | 74.2% |
| SWE Atlas v1.1: Test Writing | Scale AI SEAL | – | 58.1% |
| SWE Atlas v1.1: Refactoring | Scale AI SEAL | – | 53.3% |
| SWE-Bench Pro V2 | Scale AI SEAL | – | 97.7% |
| SWE-Bench Pro V2 HARD | Scale AI SEAL | – | 86.3% |
| CursorBench | Cursor | 47.7% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1622 | – |
| WeirdML v3 | WeirdML | 9.2%† | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 33.8% | 13.1% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 81.3% |
| GameBench 2 Overall | DuelLab GameBench 2 | 61.4† | – |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 69.9 | – |
| Design Arena Website | Design Arena | 1295† | 1335 |
| FrontierCode 1.1 (AA) | Artificial Analysis | 43.7% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 28.5% | 25.4% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 38.5% | 13.0%† |
| APEX-Agents | Mercor | 69.8% | 46.8% |
| Arena Agent | Arena (LMArena) | 2 | 2.5 |
| LiveBench Long Agentic | LiveBench | 38.8% | – |
| LiveBench Browser | LiveBench | 67.2% | – |
| MCP Atlas | Scale AI SEAL | – | 85.4% |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 62.3% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1746 | 1630 |
| τ³-Banking (AA) | Artificial Analysis | – | 40.4% |
| Analyst Agent (AA) | Artificial Analysis | – | 51.3% |
| CyberBench | Vals AI | – | 72.7% |
| Finance Agent v2 | Vals AI | 57.0% | – |
| Finance Agent v2.1 | Vals AI | 62.9% | 67.5% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 14.2% | 21.7% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 9.2% | – |
| Legal Research Bench | Vals AI | 45.2% | 46.6% |
| SREBench | Vals AI | – | 1.9% |
| Tax Agent Bench | Vals AI | 44.4% | 38.9% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 31.8% | 12.6% |
| Terminal-Bench Science | Vals AI | – | 2.9% |
| Arena Agent Task Completion | Arena (LMArena) | 2.3 | 3.3 |
| AutomationBench | Artificial Analysis | 66.6% | 45.6% |
| GDP.pdf (AA) | Artificial Analysis | 17.2% | 25.2% |
| EnterpriseOps-Gym | Artificial Analysis | – | 46.8% |
| AA-Briefcase v1.1 | Artificial Analysis | 1570 | 1527 |
| GDPx | GDPx by SemiAnalysis | 5.4% | – |
| Design Arena Agentic | Design Arena | 1282† | 1315 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 12.9% | 5.7% |
| CWE-Bench (AA) | Artificial Analysis | 69.2% | – |
| DeepSec-Bench (AA) | Artificial Analysis | 24.9% | – |
| CyberGym end-to-end (AA) | Artificial Analysis | 76.3% | – |
| MKA (AA) | Artificial Analysis | 38.5% | – |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 71.2% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 46.3% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 99.4% | – |
| ArXivMath | MathArena | 42.5% | 31.9% |
| BrokenArXiv | MathArena | 48.7% | 63.2% |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 42.5% | – |
| LiveBench 2.0 Data Analysis | LiveBench | 79.5% | – |
| LiveBench Research | LiveBench | 78.6% | – |
| AA-Omniscience v1.1 | Artificial Analysis | 0.9 | 16.1 |
| PRBench Finance | Scale AI SEAL | 47.1%† | – |
| PRBench Legal | Scale AI SEAL | 47.3%† | – |
| BioMysteryBench | Vals AI | – | 70.0% |
| Excel Modeling Benchmark | Vals AI | 61.8% | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 69.2% | 69.5% |
| Access to Justice Benchmark | Vals AI | 73.1% | – |
| DrugDiscoveryBench | Scale AI SEAL | 60.6%† | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 39.6% | 33.8% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 35.9% | 73.3% |
| Instruction following | |||
| MultiChallenge | Scale AI SEAL | 74.9%† | – |
| Arena Instruction Following | Arena (LMArena) | – | 1410 |
| Human preference | |||
| Arena Text | Arena (LMArena) | – | 1412 |
| Arena Creative Writing | Arena (LMArena) | – | 1345 |
| Arena Multi-turn | Arena (LMArena) | – | 1415 |
| Arena Longer Queries | Arena (LMArena) | – | 1410 |
| Multimodal | |||
| LiveBench Design | LiveBench | 79.8% | – |
| MMMU-Pro | Artificial Analysis | 80.6% | 81.6% |
| SAGE | Vals AI | – | 88.6% |
| Humanity's Sixth Sense | Scale AI SEAL | 30.1% | 33.3% |
| VIALS-AA | Handshake AI | 40.4% | 42.9% |
| Code2Video-AA | Artificial Analysis | – | 1020 |
| AA-3D | Artificial Analysis | – | 1027 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 82.7% | 83.3% |
| MLCR | Artificial Analysis | 26.7% | 20.6% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 159.4† | – |
| LiveBench 2.0 | LiveBench | 70.3% | – |
| AA Intelligence Index v5.0.1 | Artificial Analysis | 32.6 | 30.6 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 45.1 | – |
| AA Progress Index v1 | Artificial Analysis | 323.7 | – |
| Vals Index | Vals AI | 61.3 | 56.1 |
| Vals Finance Index | Vals AI | 60.9% | 61.5% |
| Vals Science Index | Vals AI | – | 32.7% |
| Vals Router Index | Vals AI | 61.3%† | – |
| Vals Legal Index | Vals AI | 71.9% | 73.9% |
| AA Cyber Index | Artificial Analysis | 56.8 | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 27.2%† | 16.2% |
| DistressBench | Scale AI SEAL | – | 82.3% |
| Metadata | |||
| AA Openness Index | Artificial Analysis | 44.4 | 38.9 |
Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- DeepSeek V4.1 Flash 1017 (max): 8 new results (LiveBench 2.0, LiveBench Long Agentic, LiveBench 2.0 Reasoning…).
- Inkling 2 Small: 1 score revision; Inkling 2 Small cached input price moved from $0.060/M to $0.030/M (50% down).
- DeepSeek V4.1 Flash 1017 (max): BenchLeader Index 58.5 → 58.0.
- Inkling 2 Small: 1 score revision.
- Inkling 2 Small: BenchLeader Index 58.4 → 57.9.
- DeepSeek V4.1 Flash 1017 (max): 1 new result (Terminal-Bench); 1 score revision.
- Inkling 2 Small: 2 new results (FORTRESS, Design Arena Agentic).
- DeepSeek V4.1 Flash 1017 (max): BenchLeader Index 59.0 → 58.5.
- Inkling 2 Small: 2 new results (MysteryMechanism, Vals Science Index); 1 score revision.
- DeepSeek V4.1 Flash 1017 (max): 1 score revision.
Related comparisons
- DeepSeek V4.1 Flash 1017 vs Claude Fable 5.2
- DeepSeek V4.1 Flash 1017 vs Gemini 4 Carbon
- DeepSeek V4.1 Flash 1017 vs Claude Opus 5.5
- Inkling 2 Small vs Gemini 4 Carbon
- Inkling 2 Small vs Claude Opus 5.5
- Inkling 2 Small vs GPT-6 Astra
- Inkling 2 Small vs Claude Fable 5.1
- DeepSeek V4 Pro vs Inkling 2 Small
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
DeepSeek V4.1 Flash 1017 vs Inkling 2 Small: questions
- Is DeepSeek V4.1 Flash 1017 better than Inkling 2 Small?
- DeepSeek V4.1 Flash 1017 (max) and Inkling 2 Small are neck and neck on quality (57.7 vs 57.9, within the ±4.9 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
- DeepSeek V4.1 Flash 1017 vs Inkling 2 Small: which is cheaper, and will that change?
- Inkling 2 Small is cheaper today: $0.475 against $0.525 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is DeepSeek V4.1 Flash 1017 or Inkling 2 Small better for coding?
- DeepSeek V4.1 Flash 1017 scores higher in coding (63 vs 60 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1664 for DeepSeek V4.1 Flash 1017 against 1647 for Inkling 2 Small.
- Which is faster, DeepSeek V4.1 Flash 1017 or Inkling 2 Small?
- DeepSeek V4.1 Flash 1017 streams ~1.5× faster (174 vs 100 output tokens per second, Artificial Analysis 7-day median).
- Is DeepSeek V4.1 Flash the same as DeepSeek V4.1 Flash 1017?
- DeepSeek V4.1 Flash 1017 is the checkpoint compared here (BenchLeader Index 57.7); people searching for DeepSeek V4.1 Flash land on this page because it replaced that earlier build.
More questions (5)
- Is DeepSeek V4.1 Flash 1017 or Inkling 2 Small better for agents and tool use?
- Inkling 2 Small scores higher in agentic tasks (61 vs 58 on the category index, where 50 is average). On APEX-Agents, the top shared agents & tools benchmark, it is 69.8% for DeepSeek V4.1 Flash 1017 against 46.8% for Inkling 2 Small.
- Which is better for long documents?
- They are level on long-context benchmarks (66 vs 66). Both accept 1,048,576 tokens of context.
- Which has lower latency?
- DeepSeek V4.1 Flash 1017 answers first: 39.7 s against 49.0 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on DeepSeek V4.1 Flash 1017 vs Inkling 2 Small?
- Across the 35 benchmarks both have, DeepSeek V4.1 Flash 1017 leads 16, Inkling 2 Small leads 19. Every overall index that rates both puts DeepSeek V4.1 Flash 1017 ahead.
- When was Inkling 2 Small released?
- Inkling 2 Small was released on 5 Nov 2026.