whatmodel.sh / GPT-3 ada (API base model)

GPT-3 ada (API base model)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

GPT-3 ada (API base model) is a closed-weights language model from OpenAI, released June 2020. Licence: proprietary.

Catalogue identity notes

The original OpenAI API base model 'ada' (2020-2023), retired by OpenAI in January 2024. Not the 2023 replacement babbage-002/davinci-002 and not the InstructGPT 'text-' models. OpenAI never officially published its parameter count, so params_b is omitted.

9.9±1.5comparison only: no current run option verifiedfrom 15 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of GPT-3 ada (API base model), out of 828 with a score, then more from OpenAI.

#modelscorewhereRAMreleased
814BLOOM 1.1B BigScience11±1local~2.6GB2022-07
815Pythia 410M EleutherAI10±2local~1.8GB2023-02
816GPT-3 ada (API base model) OpenAI10±2comparison only2020-06
817StableLM-Base-Alpha 7B Stability AI9±1local~35GB2023-04
818Zyte-1B venkycs (aihub-app)8±2local~2.2GB2024-01
more from OpenAI
1GPT-6 Astra OpenAI87±1api $20.00/M2026-09
3GPT-6.1 Sol OpenAI85±1api $4.00/M2026-09
7GPT-6 Sol OpenAI83±1api $4.00/M2026-09

see all LLMs · find the ones that fit your machine

$ what it's good at

math█···█··█···█ #2 of 4 * · 4.2% of its score · 1 test

Beats GPT-Neo 1.3B, GPT-2 XL (1.5B) here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
MathQA24.3% ✎█···█··█···█ beats 2/3hard · avg 46%▇▇ 4.2%too hard for it
other tests10 not in a task · 95% of its score
testresultvs the others in its rundifficultycountswhy
HellaSwag +143.4% ✎▄·█·█··▄▄··▄ beats 0/7very easy · avg 92%▇▇▇▇▇▇▇▇ 21%at its level · several runs
LAMBADA (OpenAI version), last-word accuracy +151.6% ✎█·▄··█··█··▄ beats 1/7very easy · avg 90%▇▇▇▇▇▇ 16%blunt · at its level · several runs
WinoGrande +152.9% ✎▄·▄·▄▄▄··█·▄ beats 0/7easy · avg 89%▇▇▇▇▇▇ 14%several runs
SciQ84.3% ✎██····███··█ beats 0/5very easy · avg 98%▇▇▇▇▇ 12%typical
PIQA +170.5% ✎█··▃··▃▃·▃·▃ beats 0/7easy · avg 89%▇▇▇▇ 9.2%blunt · at its level · several runs
ARC Easy51.4% ✎▄···▄··█·▄·▄ beats 0/5very easy · avg 90%▇▇▇ 8.2%at its level
TriviaQA5.0% ✎█···█·███··█ beats 0/5easy · avg 72%▇▇ 6.1%too hard for it
ARC Challenge22.5% ✎▄····▄···█▄▄ beats 0/5easy · avg 79%▇▇ 5.0%sharp · too hard for it
2 more tests
testresultvs the others in its rundifficultycountswhy
OpenBookQA17.2% ✎▅·······█··▃ beats 1/5medium · avg 52%▇ 2.7%too hard for it
PubMedQA52.8% ✎▆·▄▆·▄▂▄▄▂█▂ beats 1/20medium · avg 75%▇ 1.5%blunt · at its level

$ where to run it

No verified API or download recorded yet · add one

$ how the score was worked out · GPT-NeoX-20B: An Open-Source… 60%, EleutherAI GPT-J-6B model car… 33%, EleutherAI GPT-Neo 2.7B model… 6%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
GPT-NeoX-20B: An Open-Source Autoregressive Language Model — authors' own runs9▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 60%CC BY 4.0
EleutherAI GPT-J-6B model card — EleutherAI lm-evaluation-harness runs4▇▇▇▇▇▇▇▇▇▇ 33%Apache-2.0
EleutherAI GPT-Neo 2.7B model card — EleutherAI evaluation-harness runs2▇▇ 6%MIT
prior · pull toward the middle▇ 1%
weighted average= 9.9

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json