Noometry

Model comparison

DeepSeek-V3.1-Terminus vs Phi 3 Small 8k Instruct

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 28.8 on the Noometry Index.

Last verified . 10 shared benchmarks.

DeepSeek-V3.1-Terminus DeepSeek

42.6

Rank #96 Confirmed

Phi 3 Small 8k Instruct Microsoft

28.8

Rank #310 Confirmed

Summary

  • They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Phi 3 Small 8k Instruct in 0 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 31.3.

Side by side

DeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
ProviderDeepSeekMicrosoft
Noometry Index42.628.8
Released2025-09-222024-04-23
WeightsOpenOpen
Context window164K—
Max output66K—
Input $ / M tokens$0.30—
Output $ / M tokens$1—
Results tracked1632

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

Category by category

Coding DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 41.5 (#114), Phi 3 Small 8k Instruct: 27.4 (#313)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Coding14261102
SciCode40.6%—
LiveBench Coding—20.3%
ALE-Bench745.17—

Reasoning DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 25.7 (#126), Phi 3 Small 8k Instruct: 14.2 (#319)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Hard Prompts14271100
Kagi LLM Benchmark57.4%—
CritPt1.7%—
LiveBench Reasoning—15.9%
DTBench81.3%—
LiveBench Data Analysis—30.3%
LMCA28.6%—
Adversarial NLI—58.1%
BIG-Bench Hard—79.1%
HellaSwag—77%
LiveBench—24%
WinoGrande—81.5%

Math DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 35.7 (#138), Phi 3 Small 8k Instruct: 25.4 (#247)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Math14021151
LiveBench Math—17.6%

Knowledge Not comparable

DeepSeek-V3.1-Terminus: —, Phi 3 Small 8k Instruct: 28.5 (#237)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Expert—1067
ARC (AI2) Challenge—90.7%
MMLU—75.7%
OpenBookQA—88%
TriviaQA—58.1%

Multilingual DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 52.3 (#87), Phi 3 Small 8k Instruct: 28.6 (#267)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Non-English14071058
LMArena Russian14341111
LMArena Chinese—1061
LMArena French—1134
LMArena German—1080
LMArena Japanese—966
LMArena Korean—894
LMArena Spanish—1111

Instruction Following DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 74.4 (#96), Phi 3 Small 8k Instruct: 52.2 (#287)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Instruction Following14061087
LiveBench Instruction Following—47.2%

Long Context DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 43.5 (#88), Phi 3 Small 8k Instruct: 33.0 (#262)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Longer Query14261088

Writing & Preference DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 61.3 (#87), Phi 3 Small 8k Instruct: 31.3 (#279)

BenchmarkDeepSeek-V3.1-TerminusPhi 3 Small 8k Instruct
LMArena Text14201110
LMArena Creative Writing14041083
LMArena Multi-Turn14111068
LiveBench Language—12.9%

Frequently asked questions

Is DeepSeek-V3.1-Terminus better than Phi 3 Small 8k Instruct?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 28.8 on the Noometry Index.

Is DeepSeek-V3.1-Terminus or Phi 3 Small 8k Instruct better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 27.4 in the Noometry coding category.

How many benchmarks do DeepSeek-V3.1-Terminus and Phi 3 Small 8k Instruct share?

10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Phi 3 Small 8k Instruct has 32.

Related comparisons

Go deeper