readable AI benchmarks (simplified)
Meta
Llama 3.3 Instruct 70B
Llama 3.3 Instruct 70B has a Quality Score of 17.4, ranking 460th among 571 scored models. Per 1M tokens, pricing is $0.710 input, $0.720 output, and $0.710 cached input. It has a 128k token context window. Its Reliability Score is 22.9, ranking 435th among 573 scored models, below average (total average: 36.5). Its Value Score is 12.7, low compared with the total Value average of 31.4.
Quality Score17.4coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score12.7
Factual reliability22.9
Cache Discount0%
Model specification
- Reasoning
- non-reasoning
- Input modalities
- none
- Output modalities
- none
- Context window
- 128k tokens
- Weights
- open weights
Token prices USD per 1M tokens
- Input
- $0.710
- Output
- $0.720
- Cached Input
- $0.710
- Cache write
- not available
- Cost per task
- not available
Capability
- Intelligence
- 7.7
- Coding
- not available
- Agentic
- not available
- Omniscience
- -54.2
- Correct
- 18.9%
- Blended price
- $0.711
Answer outcomes
Fully graded outcomes (Correct + Incorrect): 92.1%
Correct 18.9%Incorrect 73.1%Partial / not attempted 7.9%
Artificial Analysis benchmarks
- GDPval-AA v2
- 0%
- Humanity’s Last Exam
- 4%
- GPQA Diamond
- 50%
- CritPt
- 0%
- AA-Omniscience
- 23%
- AA-LCR
- 16%
Similar models
- Llama 4 ScoutMetaQuality Score 18.2Value Score not available
- Llama 3.1 Instruct 70BMetaQuality Score 19.3Value Score not available
- Llama 3 Instruct 70BMetaQuality Score 15.6Value Score not available
- Llama 3.2 Instruct 1BMetaQuality Score 15.0Value Score not available
- Llama 3.2 Instruct 11B (Vision)MetaQuality Score 13.5Value Score not available
- Llama 3.1 Instruct 8BMetaQuality Score 22.6Value Score not available