whatmodel.sh / GPT-4 Turbo Preview (1106)
GPT-4 Turbo Preview (1106)
↗ share
+ contribute
GPT-4 Turbo Preview (1106) is a closed-weights language model from OpenAI, released November 2023. Licence: proprietary.
Catalogue identity notes
Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2023-11-06). Dated API snapshot gpt-4-1106-preview. OpenAI deprecations page lists it as shut down on 2026-03-26 (it also appears in the 2026-04-22 table for October 23, 2026; treated as retired).
$ how it compares
The LLMs ranked either side of GPT-4 Turbo Preview (1106), out of 828 with a score, then more from OpenAI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 341 |  Mistral Large 2 (2407) | 54±1 | local | ||
| 342 | Mistral Small 3.1 24B (2503) | 54±1 | local | ||
| 343 | GPT-4 Turbo Preview (1106) | 54±1 | comparison only | ||
| 344 | GPT-4 Turbo Preview (0125) | 54±1 | comparison only | ||
| 345 |  Claude 3.5 Haiku | 54±1 | comparison only | ||
| 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
coding▁▁▁▃▅▇█▇▅▄▃▁ #256 of 524 · 43% of its score · 6 tests
Beats Qwen3.5 4B, Qwen3-VL 32B Instruct, Solar Open 100B (102B-A12B) here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 3B Reasoning 2512, DeepSeek-Coder-V2 Instruct (0724) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| HumanEval | 85.4% ✎ | ▁·▁▁▃▄▆▅█▃▁▁ beats 90/99 | 14% |
| HumanEval+ | 81.7% ✎ | ▁·▁▁▁▃▄▄▄█▃▁ beats 92/99 | 12% |
| LiveCodeBench official (Jan 1–Nov 1, 2024) | 35.7% ✎ | ▁▁·▁▂▄▆█▅▂·▂ beats 20/31 | 7.4% |
| Aider code editing (133 Exercism Python exercises, two attempts) | 65.4% ✎ | ▁·▁▂▄▁▆▄█▆▂▂ beats 38/62 | 5.2% |
| MBPP+ | 73.3% ✎ | ▁···▁▁▂▅▇█▆▂ beats 69/75 | 2.9% |
| HumanEval (simple-evals) lab | 83.7% ✎ | ▆█▆▆▂··▂···▂ beats 0/15 | 1.1% |
chat & writing▁▁▁▃▅▄▆█▇▇▇▃ #247 of 463 · 21% of its score · 5 tests
Beats Granite 4.2 8B, Mistral Small 4 (119B-A6.5B), Gemma 2 9B IT SimPO here, though they rank higher overall. Trails LFM2.5 1.2B Instruct, Granite 4.0 H Micro 3B, Gemma 3n E4B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| AlpacaEval 2 length-controlled | 50.0% ✎ | ▁▂▂▃▆▆█▆▄▄▁▁ beats 105/122 | 13% |
| Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024) | 1265.0 ✎ | ▂▁▂▁▃▇█▆▇▄▂▃ beats 92/95 | 2.9% |
| LMArena (text) | 1312.8 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 141/379 | 2.9% |
| MT-Bench (GPT-4 single-answer grading, LMSYS runs) | 9.3200 MT-Bench score (1-10) ✎ | ▆▂▂▆▆▄██▂▄▄▆ beats 28/28 | 2.6% |
| AlpacaEval 1.0 win rate (vs text-davinci-003, GPT-4 judge) | 97.7% ✎ | ▂▂·▆·▇▃▆█▆▄▃ beats 45/45 | 0.2% |
math▁▂▃▅▇▅▅█▄▁▁▁ #316 of 462 · 10% of its score · 3 tests
Beats Molmo 2 8B here, though they rank higher overall. Trails DeepSeek R1 Distill Qwen 1.5B, Granite 8B Code Base (4K), Mixtral 8x7B v0.1 (base) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| MATH level 5 | 40.0% ✎ | ▂▂▅▆▄█▆▃▅▂▃▄ beats 32/92 | 7.3% |
| MATH (simple-evals, original test set, pre-o1 rows) lab | 64.3% ✎ | █····▄··▄··█ beats 0/5 | 1.4% |
| MGSM (simple-evals) lab | 87.1% ✎ | ▁····▁▅▁█▁▃▁ beats 5/16 | 1.1% |
other tests7 not in a task · 26% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LiveBench 2024-07-26 | 52.0% ✎ | ▁▁▁▂▂▃▂█▆▃▅▃ beats 54/61 | 13% |
| GPQA Diamond (simple-evals) lab | 42.5% ✎ | █▃▃█·▃▃▅▃▃▃▅ beats 2/16 | 5.5% |
| GPQA Diamond | 42.4% ✎ | ▁▁▁▁▂▄▆▄▅▇█▅ beats 46/208 | 4.4% |
| MMLU (simple-evals, zero-shot CoT) lab | 84.7% ✎ | ▂▂·█▄▄▂·█▂·▂ beats 2/16 | 2.2% |
| RULER 64k | 87.0% ✎ | ▁▁·▁·▂▃▄█▃·▁ beats 20/22 | 0.5% |
| DROP F1, 3-shot (simple-evals) lab | 83.2% ✎ | ▂▂▄▆▆█·▂···▄ beats 10/16 | 0.3% |
| RULER 128k | 81.2% ✎ | ▄··▂▄▂▆██▄▄▂ beats 17/21 | 0.2% |
$ where to run it
No verified API or download recorded yet · add one
$ how the score was worked out · EvalPlus / EvalPerf published… 29%, AlpacaEval 1.0 and 2.0 leader… 13%, LiveBench model judgments (HF… 13% and 8 more
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
$ 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