Compare · ModelsLive · 2 picked · head to head
Falcon-180B vs Llama 2-13B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Falcon-180B wins on 9/10 benchmarks
Falcon-180B wins 9 of 10 shared benchmarks. Leads in knowledge · math.
Category leads
knowledge·Falcon-180Breasoning·Llama 2-13Bmath·Falcon-180B
Hype vs Reality
Attention vs performance
Falcon-180B
#119 by perf·no signal
Llama 2-13B
#128 by perf·no signal
Vendor risk
Who is behind the model
TII
private · undisclosed
Meta AI
$1.50T·Tier 1
Head to head
10 benchmarks · 2 models
Falcon-180BLlama 2-13B
ARC AI2
Falcon-180B leads by +10.0
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
Falcon-180B
57.1
Llama 2-13B
47.1
BBH
Llama 2-13B leads by +28.1
BIG-Bench Hard · a curated subset of 23 challenging tasks from BIG-Bench where language models previously failed to outperform average humans.
Falcon-180B
16.1
Llama 2-13B
44.3
GSM8K
Falcon-180B leads by +17.5
Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve.
Falcon-180B
54.4
Llama 2-13B
36.9
HellaSwag
Falcon-180B leads by +11.1
HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios.
Falcon-180B
85.3
Llama 2-13B
74.3
LAMBADA
Falcon-180B leads by +3.3
LAMBADA · measures the ability to predict the final word of a passage, requiring broad contextual understanding across long text spans.
Falcon-180B
79.8
Llama 2-13B
76.5
MMLU
Falcon-180B leads by +20.0
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Falcon-180B
60.8
Llama 2-13B
40.8
OpenBookQA
Falcon-180B leads by +9.6
OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting.
Falcon-180B
52.3
Llama 2-13B
42.7
PIQA
Falcon-180B leads by +8.2
PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks.
Falcon-180B
69.8
Llama 2-13B
61.6
TriviaQA
Falcon-180B leads by +0.3
TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents.
Falcon-180B
79.9
Llama 2-13B
79.6
Winogrande
Falcon-180B leads by +28.6
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Falcon-180B
74.2
Llama 2-13B
45.6
Full benchmark table
| Benchmark | Falcon-180B | Llama 2-13B |
|---|---|---|
ARC AI2 AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval. | 57.1 | 47.1 |
BBH BIG-Bench Hard · a curated subset of 23 challenging tasks from BIG-Bench where language models previously failed to outperform average humans. | 16.1 | 44.3 |
GSM8K Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve. | 54.4 | 36.9 |
HellaSwag HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios. | 85.3 | 74.3 |
LAMBADA LAMBADA · measures the ability to predict the final word of a passage, requiring broad contextual understanding across long text spans. | 79.8 | 76.5 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 60.8 | 40.8 |
OpenBookQA OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting. | 52.3 | 42.7 |
PIQA PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks. | 69.8 | 61.6 |
TriviaQA TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents. | 79.9 | 79.6 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 74.2 | 45.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| — | — | — | — |