whatmodel.sh / GPT-3.5 text-davinci-003
GPT-3.5 text-davinci-003
↗ share
+ contribute
GPT-3.5 text-davinci-003 is a closed-weights language model from OpenAI, released November 2022. Licence: proprietary.
Catalogue identity notes
Comparison-only. OpenAI completions model text-davinci-003. OpenAI's deprecations page lists its shutdown on 2024-01-04. Release month from the snapshot date in the model name (text-davinci-002: March 2022 API release; text-davinci-003: 2022-11-28).
$ how it compares
The LLMs ranked either side of GPT-3.5 text-davinci-003, out of 1132 with a score, then more from OpenAI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 980 |  MedReason-8B | 34±5 | local | ||
| 981 |  Llama 2 13B (base) | 34±1 | local | ||
| 982 | GPT-3.5 text-davinci-003 | 34±3 | comparison only | ||
| 983 | Python-Code-13B | 34±1 | local | ||
| 984 | Code Llama 7B (base) | 34±1 | local | ||
| more from OpenAI | |||||
| 1 | GPT-6 Astra | 87±1 | api $20.00/M | ||
| 3 | GPT-6.1 Sol | 85±1 | api $4.00/M | ||
| 7 | GPT-6 Sol | 83±1 | api $4.00/M | ||
$ what it's good at
chat & writing▁▁▃▃▅▅▆█▆▆▆▂ #562 of 579 * · 52% of its score · 1 test
Beats Phi-2 (2.7B) here, though they rank higher overall. Trails Gemma 3 270M IT, SmolLM2 360M Instruct, Granite 4.0 350M here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| AlpacaEval 1.0 win rate (vs text-davinci-003, GPT-4 judge) | 50.0% ✎ | ▂▂·▆·▇▃▆█▆▄▂ beats 9/45 | 52% |
other tests1 not in a task · 43% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| EQ-Bench v1 (60 questions, paper run) | 43.7% ✎ | ▂··▄·▂▆▄▂█·▂ beats 7/14 | 43% |
$ where to run it
No verified API or download recorded yet · add one
$ how the score was worked out · AlpacaEval 1.0 and 2.0 leader… 52%, EQ-Bench: An Emotional Intell… 43%
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
| source | results | share of the score |
|---|---|---|
 AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository) | 1 | 52% |
 EQ-Bench: An Emotional Intelligence Benchmark for Large Language Models — authors' own runs | 1 | 43% |
| prior · pull toward the middle | 5% | |
| weighted average | = 34.0 |
$ improve this page
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