Noometry

Model comparison

DeepSeek-V3.1-Terminus vs Solar Pro4

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 41.5 on the Noometry Index.

Last verified . 10 shared benchmarks.

DeepSeek-V3.1-Terminus DeepSeek

42.6

Rank #96 Confirmed

Solar Pro4 Upstage

41.5

Rank #123 Confirmed

Summary

  • They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 5 categories and Solar Pro4 in 2 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 56.7.
  • DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $0.30 / $1.20 for Solar Pro4.
  • Solar Pro4 accepts more context: 524K tokens versus 164K.
  • DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.1-TerminusSolar Pro4
ProviderDeepSeekUpstage
Noometry Index42.641.5
Released2025-09-222026-08-06
WeightsOpenProprietary
Context window164K524K
Max output66K131K
Input $ / M tokens$0.30$0.30
Output $ / M tokens$1$1.20
Results tracked1618

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), Solar Pro4: 39.9 (#135)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Coding14261439
LMArena WebDev—1371
SciCode40.6%—
ALE-Bench745.17—

Reasoning Solar Pro4 leads

DeepSeek-V3.1-Terminus: 25.7 (#126), Solar Pro4: 27.3 (#107)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Hard Prompts14271398
Kagi LLM Benchmark57.4%—
CritPt1.7%—
DTBench81.3%—
LMCA28.6%—

Math Too close to call

DeepSeek-V3.1-Terminus: 35.7 (#138), Solar Pro4: 36.1 (#129)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Math14021418

Knowledge Not comparable

DeepSeek-V3.1-Terminus: —, Solar Pro4: 39.0 (#128)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Expert—1421

Multilingual DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 52.3 (#87), Solar Pro4: 49.0 (#134)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Non-English14071362
LMArena Russian14341360
LMArena Chinese—1417
LMArena French—1403
LMArena German—1351
LMArena Japanese—1300
LMArena Korean—1383
LMArena Spanish—1408

Instruction Following DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 74.4 (#96), Solar Pro4: 72.8 (#128)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Instruction Following14061376

Long Context DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 43.5 (#88), Solar Pro4: 42.0 (#125)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Longer Query14261379

Writing & Preference DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 61.3 (#87), Solar Pro4: 56.7 (#133)

BenchmarkDeepSeek-V3.1-TerminusSolar Pro4
LMArena Text14201386
LMArena Creative Writing14041313
LMArena Multi-Turn14111391

Frequently asked questions

Is DeepSeek-V3.1-Terminus better than Solar Pro4?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 41.5 on the Noometry Index.

Which is cheaper, DeepSeek-V3.1-Terminus or Solar Pro4?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; Solar Pro4 lists at $0.30 and $1.20.

Is DeepSeek-V3.1-Terminus or Solar Pro4 better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 39.9 in the Noometry coding category.

Which has the bigger context window?

Solar Pro4 does, with 524K tokens against 164K.

How many benchmarks do DeepSeek-V3.1-Terminus and Solar Pro4 share?

10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Solar Pro4 has 18.

Related comparisons

Go deeper