Noometry

Model comparison

Claude 3 Haiku vs DeepSeek-V3.1-Terminus

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 25.7 on the Noometry Index.

Last verified . 13 shared benchmarks.

Claude 3 Haiku Anthropic

25.7

Rank #333 Confirmed

DeepSeek-V3.1-Terminus DeepSeek

42.6

Rank #96 Confirmed

Summary

  • They share 13 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and DeepSeek-V3.1-Terminus in 7 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 29.8.
  • The biggest single-benchmark swing is DTBench: 50.1% for Claude 3 Haiku and 81.3% for DeepSeek-V3.1-Terminus.
  • DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuDeepSeek-V3.1-Terminus
ProviderAnthropicDeepSeek
Noometry Index25.742.6
Released2024-03-072025-09-22
WeightsProprietaryOpen
Context window—164K
Max output—66K
Input $ / M tokens—$0.30
Output $ / M tokens—$1
Results tracked3716

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 26.1 (#320), DeepSeek-V3.1-Terminus: 41.5 (#114)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Coding11991426
SciCode—40.6%
WeirdML9.8%—
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—745.17
HumanEval+68.9%—
MBPP+68.8%—

Reasoning DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 15.7 (#303), DeepSeek-V3.1-Terminus: 25.7 (#126)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
Kagi LLM Benchmark34.2%57.4%
LMArena Hard Prompts11741427
DTBench50.1%81.3%
LMCA8.8%28.6%
CritPt—1.7%
Epoch Capabilities Index118.3—
ForecastBench53.2—
WinoGrande74.2%—

Math DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 9.2 (#313), DeepSeek-V3.1-Terminus: 35.7 (#138)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Math11881402
OTIS Mock AIME 2024-20251.8%—
MATH Level 514.9%—

Knowledge Not comparable

Claude 3 Haiku: 17.3 (#280), DeepSeek-V3.1-Terminus: —

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
GPQA Diamond36.3%—
Confabulations34.2%—
LMArena Expert1148—
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.1 (#123), DeepSeek-V3.1-Terminus: —

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Vision950—
ScienceQA72%—

Multilingual DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 36.1 (#238), DeepSeek-V3.1-Terminus: 52.3 (#87)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Non-English11781407
LMArena Russian12041434
LMArena Chinese1155—
LMArena French1194—
LMArena German1174—
LMArena Japanese1102—
LMArena Korean1108—
LMArena Spanish1165—

Instruction Following DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 61.5 (#242), DeepSeek-V3.1-Terminus: 74.4 (#96)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Instruction Following11741406

Long Context DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 36.1 (#232), DeepSeek-V3.1-Terminus: 43.5 (#88)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Longer Query11911426

Writing & Preference DeepSeek-V3.1-Terminus leads

Claude 3 Haiku: 29.8 (#286), DeepSeek-V3.1-Terminus: 61.3 (#87)

BenchmarkClaude 3 HaikuDeepSeek-V3.1-Terminus
LMArena Text11951420
LMArena Creative Writing11571404
LMArena Multi-Turn11901411
EQ-Bench Creative Writing717—

Frequently asked questions

Is Claude 3 Haiku better than DeepSeek-V3.1-Terminus?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 25.7 on the Noometry Index.

Is Claude 3 Haiku or DeepSeek-V3.1-Terminus better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 26.1 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and DeepSeek-V3.1-Terminus share?

13 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.

Related comparisons

Go deeper