Compare · ModelsLive · 2 picked · head to head
open_llama_7b vs vicuna-13b-v1.1
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
vicuna-13b-v1.1 wins on 3/5 benchmarks
vicuna-13b-v1.1 wins 3 of 5 shared benchmarks. Leads in knowledge.
Category leads
knowledge·vicuna-13b-v1.1
Hype vs Reality
Attention vs performance
open_llama_7b
#214 by perf·no signal
vicuna-13b-v1.1
#212 by perf·no signal
Vendor risk
Who is behind the model
Meta AI
$1.50T·Tier 1
U
Unknown
private · undisclosed
Head to head
5 benchmarks · 2 models
open_llama_7bvicuna-13b-v1.1
ARC AI2
vicuna-13b-v1.1 leads by +6.0
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
open_llama_7b
18.3
vicuna-13b-v1.1
24.3
HellaSwag
open_llama_7b leads by +18.7
HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios.
open_llama_7b
62.4
vicuna-13b-v1.1
43.7
OpenBookQA
open_llama_7b leads by +8.0
OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting.
open_llama_7b
18.7
vicuna-13b-v1.1
10.7
PIQA
vicuna-13b-v1.1 leads by +2.8
PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks.
open_llama_7b
52.0
vicuna-13b-v1.1
54.8
Winogrande
vicuna-13b-v1.1 leads by +7.6
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
open_llama_7b
34.0
vicuna-13b-v1.1
41.6
Full benchmark table
| Benchmark | open_llama_7b | vicuna-13b-v1.1 |
|---|---|---|
ARC AI2 AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval. | 18.3 | 24.3 |
HellaSwag HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios. | 62.4 | 43.7 |
OpenBookQA OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting. | 18.7 | 10.7 |
PIQA PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks. | 52.0 | 54.8 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 34.0 | 41.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
U vicuna-13b-v1.1 | — | — | — | — |