[whatmodel.sh](../../README.md) / Yi-Lightning # Yi-Lightning 01.AI · released 2024-10 · closed weights **score 56.1** ±1.2 · #258 of 732 · from 4 benchmark results across 4 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks Yi-Lightning among every model on that task or test, worst to best. 
task            vs the others          rank  counts  tests
coding          ▁▁▁▃▅▇●▇▅▄▃▁   #233 of 469*     54%      2
chat & writing  ▁▁▁▃▆▄▆●▆▇▇▃   #189 of 402*     46%      2
coding · #233 of 469 · 54% of its score Beats [Qwen3.5 4B](qwen3-5-4b.md), [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Solar Open 100B (102B-A12B)](solar-open-100b.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 3B Reasoning 2512](ministral-3-3b-reasoning.md), [GPT-4 (0613)](gpt-4-0613.md) here, though they rank lower overall. 
test                                result  vs the others in its run  difficulty            counts  why
Aider polyglot (225 exercises, tw…   12.9%  ▂·▂●▅▂▃█▆▄▂▂ beats 5/49   very hard · avg 7.6%     28%  sharp
Aider code editing (133 Exercism …   65.4%  ▁·▁▂▄▁▆▄●▆▂▂ beats 34/62  medium · avg 51%         26%  sharp · at its level
chat & writing · #189 of 402 · 46% of its score Beats [Seed-OSS 36B Instruct](seed-oss-36b.md), [Gemma 2 9B IT SimPO](gemma-2-9b-simpo.md), [Nemotron-Cascade-14B-Thinking](nemotron-cascade-14b-thinking.md) here, though they rank higher overall. Trails [LFM2.5 1.2B Instruct](lfm2-5-1-2b-instruct.md), [LFM2.5 1.2B Thinking](lfm2-5-1-2b-thinking.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md) here, though they rank lower overall. 
test                                result  vs the others in its run    difficulty        counts  why
Arena-Hard (v0.1, GPT-4-1106 judg…   81.5%  ▂▁▁▅█▅▄▃▃▅●▁ beats 54/60    medium · avg 47%     30%  sharp · at its level
LMArena (text)                      1328.3  ▁▁▁▂▃▃▃●▆▇█▃ beats 167/385  avg 1260             15%  blunt
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's 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 is added up by who ran the tests. [Method](../method.md) 
source                             results  share of the score
Arena-Hard-Auto v0.1 leaderboard         1   30% ▇▇▇▇
Aider polyglot benchmark runs            1   28% ▇▇▇
Aider code editing benchmark runs        1   26% ▇▇▇
LMArena leaderboard dataset              1   15% ▇▇
prior: pull toward the middle                 1% ▇
weighted average                            = 56.1
Licences: Arena-Hard-Auto v0.1 leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-v0-1-Apache-2.0.txt)) · Aider polyglot benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · Aider code editing benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) ## How to run it ### Through an API api (01.AI (Lingyiwanwu platform)) · price unknown · [platform.lingyiwanwu.com](https://platform.lingyiwanwu.com/docs) [context source](https://platform.lingyiwanwu.com/docs): 01.AI platform model table: yi-lightning context length 16K (stated rounded). The same row now describes the yi-lightning endpoint as routing requests to DeepSeek-V3, Qwen3-30B-A3B, Yi-Lightning and others, so this is the 01.AI service limit for the yi-lightning API name rather than a statement about the original model's architecture. Observed 2026-10-07. [interactive version](https://whatmodel.sh/model/yi-lightning) · [JSON](https://whatmodel.sh/api?format=json&all=1)