Model comparison
DeepSeek LLM 67B vs DeepSeek-V3.1-Terminus
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 24.9 on the Noometry Index.
Last verified . 9 shared benchmarks.
DeepSeekSummary
- They share 9 benchmarks with published results for both. DeepSeek LLM 67B 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 math, where DeepSeek-V3.1-Terminus leads 38.5 to 8.7.
Side by side
| DeepSeek LLM 67B | DeepSeek-V3.1-Terminus | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 24.9 | 43.1 |
| Released | 2023-11-29 | 2025-09-22 |
| Weights | Open | Open |
| Context window | — | 164K |
| Max output | — | 147K |
| Input $ / M tokens | — | $0.27 |
| Output $ / M tokens | — | $1 |
| Results tracked | 15 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 31.9 (#278), DeepSeek-V3.1-Terminus: 42.0 (#113)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Coding | 1096 | 1426 |
| SciCode | — | 40.6% |
| ALE-Bench | — | 745.17 |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 16.5 (#304), DeepSeek-V3.1-Terminus: 26.4 (#133)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1426 |
| Kagi LLM Benchmark | — | 57.4% |
| CritPt | — | 1.7% |
| Chess Puzzles | 0% | — |
| DTBench | — | 81.3% |
| LMCA | — | 28.6% |
| Epoch Capabilities Index | 110.5 | — |
Math DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 8.7 (#324), DeepSeek-V3.1-Terminus: 38.5 (#137)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Math | 1108 | 1402 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| MATH Level 5 | 6.4% | — |
Knowledge Not comparable
DeepSeek LLM 67B: 7.0 (#313), DeepSeek-V3.1-Terminus: —
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| GPQA Diamond | 24.6% | — |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 29.4 (#267), DeepSeek-V3.1-Terminus: 52.1 (#92)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Non-English | 1073 | 1407 |
| LMArena Chinese | 1132 | — |
| LMArena Russian | — | 1436 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 55.4 (#277), DeepSeek-V3.1-Terminus: 74.0 (#106)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Instruction Following | 1079 | 1404 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 33.1 (#265), DeepSeek-V3.1-Terminus: 43.4 (#97)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Longer Query | 1092 | 1421 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek LLM 67B: 31.6 (#282), DeepSeek-V3.1-Terminus: 61.0 (#92)
| Benchmark | DeepSeek LLM 67B | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Text | 1105 | 1419 |
| LMArena Creative Writing | 1067 | 1403 |
| LMArena Multi-Turn | 1082 | 1411 |
Frequently asked questions
Is DeepSeek LLM 67B better than DeepSeek-V3.1-Terminus?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or DeepSeek-V3.1-Terminus better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and DeepSeek-V3.1-Terminus share?
9 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.