Model comparison
Codellama 70b Instruct vs DeepSeek-V3.1-Terminus
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 33.7 on the Noometry Index.
Last verified . 4 shared benchmarks.
Meta
DeepSeekSummary
- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 0 categories and DeepSeek-V3.1-Terminus in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.0 to 33.4.
Side by side
| Codellama 70b Instruct | DeepSeek-V3.1-Terminus | |
|---|---|---|
| Provider | Meta | DeepSeek |
| Noometry Index | 33.7 | 43.1 |
| Released | — | 2025-09-22 |
| Weights | Open | Open |
| Context window | — | 164K |
| Max output | — | 147K |
| Input $ / M tokens | — | $0.27 |
| Output $ / M tokens | — | $1 |
| Results tracked | 7 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
Codellama 70b Instruct: 37.6 (#193), DeepSeek-V3.1-Terminus: 42.0 (#113)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| SciCode | — | 40.6% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1426 |
| BigCodeBench Complete | 49.6% | — |
| ALE-Bench | — | 745.17 |
| HumanEval+ | 65.9% | — |
Reasoning DeepSeek-V3.1-Terminus leads
Codellama 70b Instruct: 20.1 (#242), DeepSeek-V3.1-Terminus: 26.4 (#133)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1426 |
| Kagi LLM Benchmark | — | 57.4% |
| CritPt | — | 1.7% |
| DTBench | — | 81.3% |
| LMCA | — | 28.6% |
Math Not comparable
Codellama 70b Instruct: —, DeepSeek-V3.1-Terminus: 38.5 (#137)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Math | — | 1402 |
Multilingual DeepSeek-V3.1-Terminus leads
Codellama 70b Instruct: 24.8 (#288), DeepSeek-V3.1-Terminus: 52.1 (#92)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Non-English | 992 | 1407 |
| LMArena Russian | — | 1436 |
Instruction Following DeepSeek-V3.1-Terminus leads
Codellama 70b Instruct: 51.9 (#293), DeepSeek-V3.1-Terminus: 74.0 (#106)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Instruction Following | 1024 | 1404 |
Long Context Not comparable
Codellama 70b Instruct: —, DeepSeek-V3.1-Terminus: 43.4 (#97)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Longer Query | — | 1421 |
Writing & Preference DeepSeek-V3.1-Terminus leads
Codellama 70b Instruct: 33.4 (#277), DeepSeek-V3.1-Terminus: 61.0 (#92)
| Benchmark | Codellama 70b Instruct | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Text | 1057 | 1419 |
| LMArena Creative Writing | — | 1403 |
| LMArena Multi-Turn | — | 1411 |
Frequently asked questions
Is Codellama 70b Instruct better than DeepSeek-V3.1-Terminus?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 33.7 on the Noometry Index.
Is Codellama 70b Instruct or DeepSeek-V3.1-Terminus better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 37.6 in the Noometry coding category.
How many benchmarks do Codellama 70b Instruct and DeepSeek-V3.1-Terminus share?
4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.