DeepSeek, open weights

# DeepSeek-V3

> DeepSeek-V3 by DeepSeek, released December 2024. Ranked #166 of 354 with a Noometry Index of 39.5. API: $0.24 in / $0.90 out per M tokens. 164K context. Scores, sources and comparisons.
- Canonical page: https://noometry.com/models/deepseek-v3
- Last updated: 2026-10-10
- Title: DeepSeek-V3 Benchmarks, Price & Rank (October 2026)

DeepSeek-V3 by DeepSeek ranks 166th of 354 ranked models on the Noometry Index as of October 2026, with a score of 39.5. Its strongest category is coding, where it ranks 106th. API pricing starts at $0.24 per million input tokens and $0.90 per million output tokens, with a 164K-token context window.

Last verified October 10, 2026

## Specifications

- **Noometry rank:** #166 of 354
- **Index score:** 39.5
- **Evidence:** Confirmed 60 results
- **Provider:** [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek)
- **Released:** December 26, 2024
- **Weights:** Open weights
- **Reasoning:** No
- **Context window:** 164K
- **Max output:** 164K
- **Input price:** $0.24 / M
- **Output price:** $0.90 / M
- **Blended price:** $0.41 / M
- **Output speed:** 73 tokens/s [Kagi](https://help.kagi.com/kagi/ai/llm-benchmark.html)
- **Value:** #62 of 219
- **Knowledge cutoff:** July 2024
- **Input:** text
- **Hugging Face:** [deepseek-ai/DeepSeek-V3](https://huggingface.co/deepseek-ai/DeepSeek-V3)

## Category scores

Each category score combines every public result we have in that category.

DeepSeek-V3 category scores

1.  Coding 42.3
2.  Reasoning 20.5
3.  Math 32.1
4.  Knowledge 37.5
5.  Multilingual 48.5
6.  Instruction Following 72.8
7.  Long Context 34.0
8.  Writing & Preference 57.4
9.  020406080

DeepSeek-V3 category ranks
| Category | Score | Rank | Results |
| --- | --- | --- | --- |
| [Coding](https://noometry.com/best/coding) | 42.3 | #106 | 7 |
| [Reasoning](https://noometry.com/best/reasoning) | 20.5 | #236 | 8 |
| [Math](https://noometry.com/best/math) | 32.1 | #219 | 5 |
| [Knowledge](https://noometry.com/best/knowledge) | 37.5 | #155 | 6 |
| [Multilingual](https://noometry.com/best/multilingual) | 48.5 | #143 | 1 |
| [Instruction Following](https://noometry.com/best/instruction-following) | 72.8 | #130 | 3 |
| [Long Context](https://noometry.com/best/long-context) | 34.0 | #253 | 2 |
| [Writing & Preference](https://noometry.com/best/writing) | 57.4 | #130 | 7 |

## Strengths and weaknesses

Categories where DeepSeek-V3 places highest and lowest among the models ranked in each, with its score against that category's median.

### Strongest categories

DeepSeek-V3: strongest categories
| Category | Score | vs median | Rank |
| --- | --- | --- | --- |
| [Coding](https://noometry.com/best/coding) | 42.3 | +3.6 | #106 of 340, top 32% |
| [Writing & Preference](https://noometry.com/best/writing) | 57.4 | +3.6 | #130 of 312, top 42% |
| [Instruction Following](https://noometry.com/best/instruction-following) | 72.8 | +1.5 | #130 of 305, top 43% |

### Weakest categories

DeepSeek-V3: weakest categories
| Category | Score | vs median | Rank |
| --- | --- | --- | --- |
| [Long Context](https://noometry.com/best/long-context) | 34.0 | −6.9 | #253 of 296, top 86% |
| [Reasoning](https://noometry.com/best/reasoning) | 20.5 | −3.1 | #236 of 350, top 68% |
| [Math](https://noometry.com/best/math) | 32.1 | −4.5 | #219 of 327, top 67% |

## Closest competitors

The models ranked just above and below DeepSeek-V3. When scores are this close, price and speed are often the better way to choose.

Models ranked closest to DeepSeek-V3
| Model | Rank | Score | Blended $/M | Speed |  |
| --- | --- | --- | --- | --- | --- |
| [DeepSeek-V3.2-Speciale](https://noometry.com/models/deepseek-v3-2-speciale) | #162 | 39.7 | $0.85 | — | [Compare](https://noometry.com/compare/deepseek-v3-vs-deepseek-v3-2-speciale) |
| [Hunyuan Turbo 0110](https://noometry.com/models/hunyuan-turbo) | #163 | 39.6 | — | — | [Compare](https://noometry.com/compare/deepseek-v3-vs-hunyuan-turbo) |
| [Claude 3.7 Sonnet](https://noometry.com/models/claude-3-7-sonnet) | #164 | 39.5 | — | — | [Compare](https://noometry.com/compare/claude-3-7-sonnet-vs-deepseek-v3) |
| [Claude Haiku 4.5](https://noometry.com/models/claude-haiku-4-5) | #165 | 39.5 | $2 | — | [Compare](https://noometry.com/compare/claude-haiku-4-5-vs-deepseek-v3) |
| [Grok 4 Fast](https://noometry.com/models/grok-4-fast) | #167 | 39.4 | — | 577 | [Compare](https://noometry.com/compare/deepseek-v3-vs-grok-4-fast) |
| [Olmo 3.1 32b Instruct](https://noometry.com/models/olmo-3-1-32b-instruct) | #168 | 39.4 | — | — | [Compare](https://noometry.com/compare/deepseek-v3-vs-olmo-3-1-32b-instruct) |
| [Granite 4.2 3b](https://noometry.com/models/granite-4-2-3b) | #169 | 39.4 | — | — | [Compare](https://noometry.com/compare/deepseek-v3-vs-granite-4-2-3b) |
| [Gemini 2.5 Flash](https://noometry.com/models/gemini-2-5-flash) | #170 | 39.3 | $0.85 | 152 | [Compare](https://noometry.com/compare/deepseek-v3-vs-gemini-2-5-flash) |

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

### Coding

DeepSeek-V3 Coding benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [Aider Polyglot](https://noometry.com/benchmarks/aider-polyglot) | 55.1% | #18 of 44, top 41% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [Aider Polyglot](https://noometry.com/benchmarks/aider-polyglot) | 48.4% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [SciCode](https://noometry.com/benchmarks/scicode) | 35.4% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [SciCode](https://noometry.com/benchmarks/scicode) | 35.8% | #95 of 121, top 79% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [WeirdML](https://noometry.com/benchmarks/weirdml) | 36.1% | #90 of 119, top 76% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [BigCodeBench Instruct](https://noometry.com/benchmarks/bigcodebench-instruct) | 50% | #2 of 64, top 4% |  | [BigCodeBench](https://bigcode-bench.github.io/) | 2024-12-26 |
| [LiveBench Coding](https://noometry.com/benchmarks/livebench-coding) | 70.9% | #7 of 39, top 18% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Coding](https://noometry.com/benchmarks/livebench-coding) | 61.8% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMArena Coding](https://noometry.com/benchmarks/arena-coding) | 1325 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Coding](https://noometry.com/benchmarks/arena-coding) | 1368 | #159 of 294, top 55% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [BigCodeBench Complete](https://noometry.com/benchmarks/bigcodebench-complete) | 62.2% | Best of 66 |  | [BigCodeBench](https://bigcode-bench.github.io/) | 2024-12-26 |
| [HumanEval+](https://noometry.com/benchmarks/humaneval-plus) | 86.6% | #5 of 45, top 12% | nov 2024 | [EvalPlus](https://evalplus.github.io/leaderboard.html) |  |
| [MBPP+](https://noometry.com/benchmarks/mbpp-plus) | 73% | #10 of 38, top 27% | nov 2024 | [EvalPlus](https://evalplus.github.io/leaderboard.html) |  |

### Agentic & Tool Use

DeepSeek-V3 Agentic & Tool Use benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [METR Time Horizons](https://noometry.com/benchmarks/metr-time-horizons) | 47.4% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [METR Time Horizons](https://noometry.com/benchmarks/metr-time-horizons) | 49.6% | #24 of 32, top 75% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

### Reasoning

DeepSeek-V3 Reasoning benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [SimpleBench](https://noometry.com/benchmarks/simplebench) | 18.9% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [SimpleBench](https://noometry.com/benchmarks/simplebench) | 27.2% | #62 of 77, top 81% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [Kagi LLM Benchmark](https://noometry.com/benchmarks/kagi-reasoning) | 52.3% | #58 of 99, top 59% |  | [Kagi LLM Benchmark](https://help.kagi.com/kagi/ai/llm-benchmark.html) |  |
| [CritPt](https://noometry.com/benchmarks/critpt) | 0% | #103 of 134, top 77% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [CritPt](https://noometry.com/benchmarks/critpt) | 0% | #103 of 134, top 77% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Reasoning](https://noometry.com/benchmarks/livebench-reasoning) | 65.8% | #12 of 39, top 31% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Reasoning](https://noometry.com/benchmarks/livebench-reasoning) | 56.8% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMArena Hard Prompts](https://noometry.com/benchmarks/arena-hard-prompts) | 1312 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Hard Prompts](https://noometry.com/benchmarks/arena-hard-prompts) | 1365 | #150 of 297, top 51% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [DTBench](https://noometry.com/benchmarks/dtbench) | 64.8% | #102 of 151, top 68% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Data Analysis](https://noometry.com/benchmarks/livebench-data-analysis) | 60.4% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Data Analysis](https://noometry.com/benchmarks/livebench-data-analysis) | 60.9% | #13 of 39, top 34% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMCA](https://noometry.com/benchmarks/lmca) | 15.5% | #104 of 125, top 84% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [BIG-Bench Hard](https://noometry.com/benchmarks/bbh) | 87.5% | #2 of 27, top 8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [Epoch Capabilities Index](https://noometry.com/benchmarks/epoch-capabilities-index) | 135.94 | #122 of 213, top 58% |  | [Epoch AI](https://epoch.ai/eci) | 2025-03-24 |
| [Epoch Capabilities Index](https://noometry.com/benchmarks/epoch-capabilities-index) | 132.34 |  |  | [Epoch AI](https://epoch.ai/eci) | 2024-12-26 |
| [ForecastBench](https://noometry.com/benchmarks/forecastbench) | 59.1 | #41 of 72, top 57% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [HellaSwag](https://noometry.com/benchmarks/hellaswag) | 88.9% | #5 of 29, top 18% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench](https://noometry.com/benchmarks/livebench) | 60.5% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench](https://noometry.com/benchmarks/livebench) | 66.9% | #10 of 39, top 26% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [PIQA](https://noometry.com/benchmarks/piqa) | 84.7% | #6 of 27, top 23% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [WinoGrande](https://noometry.com/benchmarks/winogrande) | 85.2% | #7 of 43, top 17% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

### Math

DeepSeek-V3 Math benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [OTIS Mock AIME 2024-2025](https://noometry.com/benchmarks/otis-mock-aime) | 37.8% | #117 of 173, top 68% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-04-01 |
| [OTIS Mock AIME 2024-2025](https://noometry.com/benchmarks/otis-mock-aime) | 15.8% |  |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-02-25 |
| [Omni-MATH](https://noometry.com/benchmarks/omni-math) | 40.3% | #26 of 57, top 46% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| [LiveBench Math](https://noometry.com/benchmarks/livebench-math) | 73.5% | #9 of 39, top 24% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Math](https://noometry.com/benchmarks/livebench-math) | 60.5% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMArena Math](https://noometry.com/benchmarks/arena-math) | 1373 | #148 of 285, top 52% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Math](https://noometry.com/benchmarks/arena-math) | 1311 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [MATH Level 5](https://noometry.com/benchmarks/math-level-5) | 64.9% |  |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-01-27 |
| [MATH Level 5](https://noometry.com/benchmarks/math-level-5) | 75.5% | #27 of 79, top 35% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-04-01 |
| [FrontierMath (Feb 2025 set)](https://noometry.com/benchmarks/frontiermath-2025-02) | 1.7% | #55 of 68, top 81% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-03-07 |

### Knowledge

DeepSeek-V3 Knowledge benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [GPQA Diamond](https://noometry.com/benchmarks/gpqa-diamond) | 67.6% | #101 of 186, top 55% |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-04-01 |
| [GPQA Diamond](https://noometry.com/benchmarks/gpqa-diamond) | 56.5% |  |  | [Epoch AI](https://epoch.ai/benchmarks) | 2025-01-27 |
| [MMLU-Pro](https://noometry.com/benchmarks/mmlu-pro) | 72.3% | #32 of 58, top 56% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| [Confabulations](https://noometry.com/benchmarks/confabulations) (lower is better) | 26.1% | #43 of 51, top 85% |  | [Lech Mazur benchmarks](https://github.com/lechmazur/confabulations) |  |
| [Vectara Hallucination Rate](https://noometry.com/benchmarks/vectara-hallucination) (lower is better) | 6.1% | #23 of 96, top 24% |  | [Vectara Hallucination Leaderboard](https://github.com/vectara/hallucination-leaderboard) |  |
| [GPQA (HELM)](https://noometry.com/benchmarks/helm-gpqa) | 53.8% | #28 of 57, top 50% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| [LMArena Expert](https://noometry.com/benchmarks/arena-expert) | 1306 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Expert](https://noometry.com/benchmarks/arena-expert) | 1351 | #156 of 273, top 58% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [ARC (AI2) Challenge](https://noometry.com/benchmarks/arc-challenge) | 95.3% | Best of 39 |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [MMLU](https://noometry.com/benchmarks/mmlu) | 87.2% | #3 of 81, top 4% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [TriviaQA](https://noometry.com/benchmarks/triviaqa) | 82.9% | #7 of 25, top 29% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

### Multilingual

DeepSeek-V3 Multilingual benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Non-English](https://noometry.com/benchmarks/arena-non-english) | 1358 | #143 of 297, top 49% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Non-English](https://noometry.com/benchmarks/arena-non-english) | 1316 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Chinese](https://noometry.com/benchmarks/arena-chinese) | 1338 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Chinese](https://noometry.com/benchmarks/arena-chinese) | 1391 | #143 of 285, top 51% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena French](https://noometry.com/benchmarks/arena-french) | 1343 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena French](https://noometry.com/benchmarks/arena-french) | 1385 | #129 of 223, top 58% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena German](https://noometry.com/benchmarks/arena-german) | 1324 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena German](https://noometry.com/benchmarks/arena-german) | 1374 | #114 of 231, top 50% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Japanese](https://noometry.com/benchmarks/arena-japanese) | 1266 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Japanese](https://noometry.com/benchmarks/arena-japanese) | 1333 | #108 of 211, top 52% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Korean](https://noometry.com/benchmarks/arena-korean) | 1319 | #120 of 213, top 57% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Korean](https://noometry.com/benchmarks/arena-korean) | 1248 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Russian](https://noometry.com/benchmarks/arena-russian) | 1373 | #133 of 283, top 47% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Russian](https://noometry.com/benchmarks/arena-russian) | 1324 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Spanish](https://noometry.com/benchmarks/arena-spanish) | 1358 | #137 of 226, top 61% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Spanish](https://noometry.com/benchmarks/arena-spanish) | 1352 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Instruction Following

DeepSeek-V3 Instruction Following benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LiveBench Instruction Following](https://noometry.com/benchmarks/livebench-if) | 75.3% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Instruction Following](https://noometry.com/benchmarks/livebench-if) | 81.5% | #8 of 39, top 21% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [IFEval](https://noometry.com/benchmarks/ifeval) | 83.2% | #30 of 57, top 53% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| [LMArena Instruction Following](https://noometry.com/benchmarks/arena-instruction-following) | 1315 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Instruction Following](https://noometry.com/benchmarks/arena-instruction-following) | 1345 | #149 of 298, top 50% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Long Context

DeepSeek-V3 Long Context benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [Fiction.LiveBench](https://noometry.com/benchmarks/fiction-livebench) | 50% | #34 of 47, top 73% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LMArena Longer Query](https://noometry.com/benchmarks/arena-longer-query) | 1352 | #152 of 291, top 53% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Longer Query](https://noometry.com/benchmarks/arena-longer-query) | 1343 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |

### Writing & Preference

DeepSeek-V3 Writing & Preference benchmark results
| Benchmark | Score | Position | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [LMArena Text](https://noometry.com/benchmarks/arena-text) | 1333 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Text](https://noometry.com/benchmarks/arena-text) | 1375 | #141 of 297, top 48% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Creative Writing](https://noometry.com/benchmarks/arena-creative-writing) | 1329 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Creative Writing](https://noometry.com/benchmarks/arena-creative-writing) | 1364 | #116 of 295, top 40% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [Short-Story Creative Writing](https://noometry.com/benchmarks/lech-mazur-writing) | 77% | #19 of 39, top 49% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [EQ-Bench Creative Writing](https://noometry.com/benchmarks/eqbench-creative-writing) | 1472 | #62 of 115, top 54% |  | [EQ-Bench](https://eqbench.com/creative_writing.html) |  |
| [WildBench](https://noometry.com/benchmarks/wildbench) | 83% | #18 of 57, top 32% |  | [HELM Capabilities](https://crfm.stanford.edu/helm/capabilities/latest/) |  |
| [LMArena Multi-Turn](https://noometry.com/benchmarks/arena-multi-turn) | 1389 | #130 of 295, top 45% |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LMArena Multi-Turn](https://noometry.com/benchmarks/arena-multi-turn) | 1349 |  |  | [LMArena](https://lmarena.ai/leaderboard/text) | 2026-10-08 |
| [LiveBench Language](https://noometry.com/benchmarks/livebench-language) | 49.1% | #12 of 39, top 31% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| [LiveBench Language](https://noometry.com/benchmarks/livebench-language) | 47.5% |  |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

## API pricing by provider

DeepSeek-V3 API prices
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
| --- | --- | --- | --- | --- |
| [deepinfra](https://deepinfra.com/models) | $0.24 | $0.90 | $0.14 | 2026-10-10 |
| [openrouter](https://openrouter.ai/deepseek/deepseek-chat) | $0.32 | $0.89 | — | 2026-10-10 |
| [together](https://docs.together.ai/docs/serverless-models) | $1.25 | $1.25 | — | 2026-10-10 |

[All DeepSeek API prices →](https://noometry.com/llm-pricing/deepseek) [Estimate your cost →](https://noometry.com/tools/cost-calculator)

## Compare DeepSeek-V3

-   [DeepSeek-V3 vs Claude Haiku 4.5](https://noometry.com/compare/claude-haiku-4-5-vs-deepseek-v3)
-   [DeepSeek-V3 vs Grok 4 Fast](https://noometry.com/compare/deepseek-v3-vs-grok-4-fast)
-   [DeepSeek-V3 vs Claude 3.7 Sonnet](https://noometry.com/compare/claude-3-7-sonnet-vs-deepseek-v3)
-   [DeepSeek-V3 vs Olmo 3.1 32b Instruct](https://noometry.com/compare/deepseek-v3-vs-olmo-3-1-32b-instruct)
-   [DeepSeek-V3 vs Hunyuan Turbo 0110](https://noometry.com/compare/deepseek-v3-vs-hunyuan-turbo)
-   [DeepSeek-V3 vs Granite 4.2 3b](https://noometry.com/compare/deepseek-v3-vs-granite-4-2-3b)
-   [DeepSeek-V3 vs GPT-6 Astra](https://noometry.com/compare/deepseek-v3-vs-gpt-6-astra)
-   [DeepSeek-V3 vs Claude Fable 5.1](https://noometry.com/compare/claude-fable-5-1-vs-deepseek-v3)
-   [DeepSeek-V3 vs Gemini 3.8 Flash](https://noometry.com/compare/deepseek-v3-vs-gemini-3-8-flash)
-   [DeepSeek-V3 vs Kimi K3](https://noometry.com/compare/deepseek-v3-vs-kimi-k3)
-   [DeepSeek-V3 vs Grok 4.6](https://noometry.com/compare/deepseek-v3-vs-grok-4-6)
-   [DeepSeek-V3 vs Qwen3.8 Max](https://noometry.com/compare/deepseek-v3-vs-qwen3-8-max)
-   [DeepSeek-V3 vs GLM-5.3](https://noometry.com/compare/deepseek-v3-vs-glm-5-3)
-   [DeepSeek-V3 vs Muse Spark 1.3](https://noometry.com/compare/deepseek-v3-vs-muse-spark-1-3)

## Other DeepSeek models

-   [DeepSeek V4 Pro](https://noometry.com/models/deepseek-v4-pro)54.3
-   [DeepSeek V4 Flash](https://noometry.com/models/deepseek-v4-flash)53.6
-   [DeepSeek V4.1 Flash](https://noometry.com/models/deepseek-v4-1-flash)52.8
-   [DeepSeek-V3.2-Exp](https://noometry.com/models/deepseek-v3-2-exp)44.3
-   [DeepSeek-V3.1-Terminus](https://noometry.com/models/deepseek-v3-1-terminus)43.1
-   [DeepSeek-V3.1](https://noometry.com/models/deepseek-v3-1)42.8
-   [DeepSeek-R1](https://noometry.com/models/deepseek-r1)42.3
-   [DeepSeek-V3.2-Speciale](https://noometry.com/models/deepseek-v3-2-speciale)39.7

## Frequently asked questions

### How good is DeepSeek-V3?

DeepSeek-V3 by DeepSeek ranks 166th of 354 ranked models on the Noometry Index as of October 2026, with a score of 39.5. Its strongest category is coding, where it ranks 106th. API pricing starts at $0.24 per million input tokens and $0.90 per million output tokens, with a 164K-token context window.

### How much does DeepSeek-V3 cost?

DeepSeek-V3 costs $0.24 per million input tokens and $0.90 per million output tokens on deepinfra, with cached input at $0.14.

### What is DeepSeek-V3's context window?

DeepSeek-V3 accepts up to 164K tokens of input and can write up to 164K tokens in one response.

### Is DeepSeek-V3 open source?

Yes. DeepSeek-V3's weights are downloadable from Hugging Face (deepseek-ai/DeepSeek-V3); check the license for commercial terms.

### How fast is DeepSeek-V3?

DeepSeek-V3 generated about 73 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

### What are DeepSeek-V3's strengths and weaknesses?

Relative to other ranked models, DeepSeek-V3 places best in coding, writing & preference, instruction following and lowest in long context, reasoning, math.

### What is DeepSeek-V3 best at?

Its best category is coding, where it ranks 106th on Noometry.

### Cite this page

Noometry. (2026). DeepSeek-V3 benchmarks and pricing. Retrieved October 10, 2026, from https://noometry.com/models/deepseek-v3

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/models/deepseek-v3.md).