Data as of October 17, 2026 · How the score is built
Released Jan 20, 2025128K context
DeepSeek-R1
Decision readingDeepSeek-R1 scores 41.1 out of 100 and ranks #134 of 224. This profile shows 8 source-displayable benchmark rows; its strongest eligible category is Instruction Following at #103. API pricing is $0.55 input and $2.19 output per million tokens, with cached input at $0.14.
Released Jan 20, 2025 — see all recent releases
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
41.1/100
field median 49.6#134 of 224 ranked models
Public
#134of 224
Verified #70 of 88
Price
$0.55input / $2.19 output
input median $0.75cached $0.14 · blended $1.37
Speed
35tok/s
field median 92 tok/sFirst token 57.8 s
Context
128Ktokens
field median 256,000Reported for this model; direct source link not stored
Strongest published evidence
Instruction Following ranks #103. A well-rounded choice across a range of tasks.
Validate before choosing
8 published rows leave some tracked benchmark slots empty. Knowledge is its lowest eligible category at #112.
Source-linked · 8 displayable benchmark rows
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| CodingRank Not rankedWeight 20%1 benchmarkVerified | 27.5 | 1 benchmark | Verified | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeRank #112 of 184Percentile 39thWeight 12%4 benchmarksVerified | 41.5 | 4 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #103 of 148Percentile 31stWeight 5%1 benchmarkVerified | 45.9 | 1 benchmark | Verified | |||
| MathWeight 5%2 benchmarksVerified | Score pending | 2 benchmarks | Verified |
8 of 681 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- AgenticNot measured
- Coding1/1 verified
- ReasoningNot measured
- MultimodalNot measured
- Knowledge4/4 verified
- MultilingualNot measured
- Inst. Following1/1 verified
- Math2/2 verified
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
DeepSeek-R1 category percentile values
- AgenticNot eligible
- CodingNot eligible
- ReasoningNot eligible
- MultimodalNot eligible
- Knowledge39th percentile
- MultilingualNot eligible
- Instruction following31st percentile
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- AgenticNot ranked
- CodingNot ranked
- ReasoningNot ranked
- MultimodalNot ranked
- Knowledge#112/184
- MultilingualNot ranked
- Inst. Following#103/148
- MathNot ranked
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score49.2% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap46.8 behind | WeightDisplay only | Provider exact |
Knowledge4 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMLU-ProMassive Multitask Language Understanding Professional | Score84% | Versus best verified row Best verified: Qwen3.7 Max · 89.6% | Gap5.6 behind | Weight6% ref. weight | Provider exact |
| GPQAGraduate-Level Google-Proof Q&A | Score71.5% | Versus best verified row Best verified: GPT-6 Astra · 96% | Gap24.5 behind | Weight3% ref. weight | Provider exact |
| MMLUMassive Multitask Language Understanding | Score90.8% | Versus best verified row Best verified: o1 · 91.8% | Gap1 behind | WeightDisplay only | Provider exact |
| SimpleQAMeasuring Short-Form Factuality in Large Language Models | Score30.1% | Versus best verified row Best verified: GPT-4.5 · 62.5% | Gap32.4 behind | WeightDisplay only | Provider exact |
Inst. Following1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFEvalInstruction-Following Eval | Score83.3% | Versus best verified row Best verified: LFM2.5-24B-A2B · 96.8% | Gap13.5 behind | WeightDisplay only | Provider exact |
Math2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| AIME 2024American Invitational Mathematics Examination 2024 | Score79.8% | Versus best verified row Best verified: GPT-OSS 120B · 95.8% | Gap16 behind | WeightDisplay only | Provider exact |
| MATH-500MATH-500 Problem Set | Score97.3% | Versus best verified row Best verified: Ternary Bonsai 2 27B · 98.8% | Gap1.5 behind | WeightDisplay only | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 8 rowsWhat it costs to get this score
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
DeepSeek-R1 · 41.1 score · $1.37 blended per million tokens
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
Published price history
This chart appears only when at least two dated first-party price records exist for the exact model. It shows rate changes, not an inferred cost trend.
2 dated first-party price records. The newest record lists $2.19 per million tokens.
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
DeepSeek-R1-0528Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not published
- Context window
- 128K
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- Not sourced yet
- Cloud regions
- Not tracked yet
- Lifecycle
- Superseded
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.14 per million cached input tokens
- Self-host
- DeepSeek-R1 needs ~640GB VRAM (8× NVIDIA H100 (80GB)).
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
DeepSeek-R1 ranks #134 of 224 on the public leaderboard with a score of 41.08/100. Its source-verified position is #70 of 88.
DeepSeek-R1 is a open weight model with a 128K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
DeepSeek-R1 sits in the DeepSeek-R1 family with DeepSeek-R1-0528. 8 of 681 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #103, while its lowest eligible position is Knowledge at #112. a well-rounded choice across a range of tasks.
Last updated October 17, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Questions
How does DeepSeek-R1 perform overall in AI benchmarks?
DeepSeek-R1 ranks #134 out of 224 models on the public BenchAlign leaderboard, with a score of 41.08/100. Its evidence status is Supported, and this profile shows 8 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is DeepSeek-R1 good for knowledge and understanding?
DeepSeek-R1 ranks #112 out of 184 eligible models for knowledge and understanding, with a public category score of 41.5/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is DeepSeek-R1 good for coding and programming?
DeepSeek-R1 has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is DeepSeek-R1 good for mathematics?
DeepSeek-R1 has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is DeepSeek-R1 good for instruction following?
DeepSeek-R1 ranks #103 out of 148 eligible models for instruction following, with a public category score of 45.9/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is DeepSeek-R1 open source?
DeepSeek-R1 is an open-weight model from DeepSeek. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Which sibling models are related to DeepSeek-R1?
DeepSeek-R1 belongs to the DeepSeek-R1 family. Related tracked variants include DeepSeek-R1-0528. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does DeepSeek-R1 have full benchmark coverage on BenchLM?
No. DeepSeek-R1 currently has 18 source-displayable rows across 681 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of DeepSeek-R1?
DeepSeek-R1 has a reported context window of 128K in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.