Model comparison
DeepSeek LLM 67B vs Falcon-180B
Falcon-180B is the stronger model overall, scoring 32.2 to 24.9 on the Noometry Index.
Last verified . 7 shared benchmarks.
DeepSeek
Technology Innovation InstituteSummary
- They share 7 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 3 categories and Falcon-180B in 1 category; 4 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where DeepSeek LLM 67B leads 29.4 to 25.2.
Side by side
| DeepSeek LLM 67B | Falcon-180B | |
|---|---|---|
| Provider | DeepSeek | Technology Innovation Institute |
| Noometry Index | 24.9 | 32.2 |
| Released | 2023-11-29 | 2023-09-06 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
DeepSeek LLM 67B: 31.9 (#278), Falcon-180B: —
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| LMArena Coding | 1096 | — |
Reasoning Falcon-180B leads
DeepSeek LLM 67B: 16.5 (#304), Falcon-180B: 19.1 (#269)
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1007 |
| Epoch Capabilities Index | 110.5 | 112.13 |
| Chess Puzzles | 0% | — |
| HellaSwag | — | 89% |
| LAMBADA | — | 79.8% |
| PIQA | — | 84.9% |
| WinoGrande | — | 87.1% |
Math Not comparable
DeepSeek LLM 67B: 8.7 (#324), Falcon-180B: —
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| LMArena Math | 1108 | — |
| MATH Level 5 | 6.4% | — |
| GSM8K | — | 54.4% |
Knowledge Not comparable
DeepSeek LLM 67B: 7.0 (#313), Falcon-180B: —
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| ARC (AI2) Challenge | — | 67.8% |
| BoolQ | — | 89% |
| MMLU | — | 70.6% |
| OpenBookQA | — | 64.2% |
Multilingual DeepSeek LLM 67B leads
DeepSeek LLM 67B: 29.4 (#267), Falcon-180B: 25.2 (#286)
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| LMArena Non-English | 1073 | 1000 |
| LMArena Chinese | 1132 | — |
Instruction Following DeepSeek LLM 67B leads
DeepSeek LLM 67B: 55.4 (#277), Falcon-180B: 53.4 (#286)
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1047 |
Long Context Not comparable
DeepSeek LLM 67B: 33.1 (#265), Falcon-180B: —
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| LMArena Longer Query | 1092 | — |
Writing & Preference DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.6 (#282), Falcon-180B: 29.1 (#295)
| Benchmark | DeepSeek LLM 67B | Falcon-180B |
|---|---|---|
| LMArena Text | 1105 | 1054 |
| LMArena Creative Writing | 1067 | 1089 |
| LMArena Multi-Turn | 1082 | 1013 |
Frequently asked questions
Is DeepSeek LLM 67B better than Falcon-180B?
Falcon-180B is the stronger model overall, scoring 32.2 to 24.9 on the Noometry Index.
How many benchmarks do DeepSeek LLM 67B and Falcon-180B share?
7 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Falcon-180B has 16.