4.7%
AA Intelligence Index v4.1
Models / DeepSeek
DeepSeek Coder V2
DeepSeek Coder · Released 2024-05-15
DeepSeek's May 2024 coding model. HumanEval 92.7% pass@1 (open weights). Baseline before V4 and R1.
Legacy open-weights coding route — prefer DeepSeek V4 or R1.
#166 of 168 on AA Index · current snapshot
5 AA Index · 5 benchmark rows · 2 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.Composite55
Context128K
Input / 1M$0.14
Output / 1M$0.28
Knowledge cutoffopen
Statuslightweight
Family profile
Best published score in each covered benchmark family.
57/100 avg
- Reasoning
- 5
- Math
- 74
- Coding
- 93
3 tested benchmark families
Benchmark placements
Per-benchmark positions
Where DeepSeek Coder V2 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| HumanEval | Coding | #11/ 15 | 93 | DeepSeek Coder V2 repository | 2024-05-15 |
| MATH | Math | #22/ 25 | 74 | Artificial Analysis | 2026-09-01 |
| AA time to first token | Performance | #101/ 186 | Artificial Analysis | 2026-09-01 | |
| AA output speed | Performance | #102/ 187 | Artificial Analysis | 2026-09-01 | |
| Artificial Analysis Intelligence Index | Reasoning | #185/ 187 | 5 | Artificial Analysis | 2026-09-01 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where DeepSeek Coder V2 shows up, it's highlighted.
Knowledge
- 191
- 291
- 391
ReasoningDeepSeek Coder V2 · 5
- 195
- 295
- 395
MathDeepSeek Coder V2 · 74
- 199
- 299
- 399
CodingDeepSeek Coder V2 · 93
- 196
- 296
- 396
Agentic
- 1100
- 2100
- 3100
Long context
- 183
- 283
- 381
Tool use
- 189
- 287
- 387
Safety
- 192
- 292
- 391
Human preference
- 195
- 292
- 391
Newest receipts
4 recentLast 120 days
Benchmark rows added to the public ledger for DeepSeek Coder V2 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 0 t/s | Artificial Analysis | 5 |
| AA time to first token | Performance | 0.00s | Artificial Analysis | 5 |
| Artificial Analysis Intelligence Index | Reasoning | 4.7% | Artificial Analysis | 5 |
| MATH | Math | 74.27% | Artificial Analysis | 5 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where DeepSeek Coder V2 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA
Full benchmark ledger
Every catalog benchmark for DeepSeek Coder V2. Scores link to the original source; gaps mean no public row exists yet.
5 of 37 catalog benchmarks have a sourced row for DeepSeek Coder V2.
Reasoning
Math
MATHVerified
74.27%
MATH-500 (AA run)
Coding
92.7% pass@1
HumanEval
Performance
AA output speedVerified
0 t/s
Median output tokens/s (1k prompt, default provider)
AA time to first tokenVerified
0.00s
Median time to first token (1k prompt, default provider)
Sources
Where the numbers come from
- DeepSeek Coder V2· checked 2026-06-09