whatmodel.sh / K-EXAONE 2.0 (750B-A37B)

K-EXAONE 2.0 (750B-A37B)

+ contribute

K-EXAONE 2.0 (750B-A37B) is an open-weights language model from LG AI Research, released July 2026. Its published size is 750B parameters (37B active per token) and a 262k-token context window. Licence: apache-2.0.

Catalogue identity notes

LG AI Research, repo LGAI-EXAONE/K-EXAONE-2.0-750B-A37B @fec0c8d (created 2026-07-29, licence apache-2.0). Printed as 'K-EXAONE 2.0' (750B / 37B) in its card and as 'K-EXAONE 2.0 (Reasoning)' in the technical report (arXiv 2608.04505v1, CC BY-NC-ND, read for setup only). FriendliAI is linked from the card as an API partner; no first-party API page verified.

Model information sources

Release date ↗ · Weights licence ↗

Model card @fec0c8d: 'Number of Parameters 750B', 'Active Parameters 37B' (safetensors metadata 749,357,484,800).

67.2±1.3#111 of 719 LLMsfrom 11 benchmark results across 1 source

$ how it compares

The LLMs ranked either side of K-EXAONE 2.0 (750B-A37B), out of 828 with a score, then more from LG AI Research.

#	model	score	where	RAM	released
125	Gemini 2.5 Pro Google	67±1	api $3.44/M		2025-06
126	Claude Opus 4 Anthropic	67±1	comparison only		2025-05
127	K-EXAONE 2.0 (750B-A37B) LG AI Research	67±1	local	~791GB	2026-07
128	Claude Opus 4.1 Anthropic	67±1	comparison only		2025-08
129	Kimi K2 Thinking Moonshot AI	67±1	api $1.07/M +local	~160GB	2025-11
	more from LG AI Research
162	K-EXAONE (236B-A23B) LG AI Research	65±1	local	~152GB	2025-12
164	EXAONE 4.5 33B LG AI Research	65±2	local	~24GB	2026-04
251	EXAONE 4.0 32B LG AI Research	59±1	local	~20GB	2025-07

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #97 of 300 · 23% of its score · 3 tests

Beats Grok 4.20 (0309 reasoning), Grok 4.3, o3-pro here, though they rank higher overall. Trails Nova 2 Pro (Preview), MiniMax-M2.1, Qwen3.5 35B-A3B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Verified lab	68.2% ✎	▁▁▁▁▁▁▂▄█▆▁▁ beats 70/131	hard · avg 20%	▇▇▇▇ 11%	sharp · at its level · small run
Terminal-Bench 2.1 lab	43.8% ✎	▁·▁·▁▁▄▃▄█▃▇ beats 16/56	very hard · avg 11%	▇▇▇ 8.6%	at its level · small run
τ³-bench Banking lab	14.2% ✎	▂▂·▅█▂▂▃▂▅▂▃ beats 13/20	very hard · avg 5.6%	▇ 2.7%	blunt · small run

coding▁▁▁▃▆██▇▆▄▃▁ #92 of 458 * · 11% of its score · 1 test

Beats Grok 4.20 (0309 reasoning), Grok 4.3, Muse Glimmer 30B here, though they rank higher overall. Trails Llama 3.1 Nemotron Ultra 253B v1, Qwen3 235B-A22B Thinking 2507 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Verified lab	68.2% ✎	▁▁▁▁▁▁▂▄█▆▁▁ beats 70/131	hard · avg 20%	▇▇▇▇ 11%	sharp · at its level · small run

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #104 of 393 * · 5.1% of its score · 1 test

Beats LongCat 2.0 (1.6T-A48B), Inkling Small (266B-A12B), GPT-5.4 nano here, though they rank higher overall. Trails Gemini 2.5 Flash, Mistral Large 3 (675B-A41B), GPT-4.1 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
IFEval (aggregation unspecified) lab	92.4% ✎	▁▁▃▁▂▄▄▆█▃▂▁ beats 125/141	easy · avg 80%	▇▇ 5.1%	typical

math▁▂▃▅▆▅▄█▄▁▁▁ #109 of 398 · 38% of its score · 3 tests

Beats Claude Sonnet 4.6, MiniMax-M3, MiMo-V2.5 (310B-A15B) here, though they rank higher overall. Trails Gemini 2.5 Flash, Qwen3-VL 32B Thinking, gpt-oss-120b here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
AIME 2026 lab	92.3% ✎	▁▁▁▂▁▂▆█▆▂·▁ beats 29/49	medium · avg 60%	▇▇▇▇▇▇ 15%	sharp · maxed out · small run
HMMT February 2026 lab	78.4% ✎	▃▃▂▃·██▆█▅▃▅ beats 17/33	hard · avg 34%	▇▇▇▇▇▇ 14%	sharp · small run
IMO-AnswerBench lab	78.6% ✎	▁▁·▁▁▁▄▅█▄▂▃ beats 18/42	medium · avg 38%	▇▇▇ 8.5%	small run

other tests4 not in a task · 33% of its score

test	result	vs the others in its run	difficulty	counts	why
GPQA Diamond lab	82.2% ✎	▁▁▁▁▂▄▅▅███▄ beats 210/310	hard · avg 43%	▇▇▇▇▇ 13%	sharp · small run
MMLU-Pro lab	83.5% ✎	▁▁·▁▁▁▃▂▄▅█▁ beats 157/188	medium · avg 56%	▇▇▇ 8.3%	small run
Humanity's Last Exam lab	18.3% ✎	▁▁▁▃▆▆▇██▆▁▃ beats 44/104	very hard · avg 4.3%	▇▇▇ 7.3%	small run
IFBench (aggregation unspecified) lab	72.6% ✎	▃▇▂▄▅▅▂▄▄▆█▇ beats 58/80	medium · avg 41%	▇▇ 4.4%	at its level · small run

$ where to run it

Other ways to run it

Run it as published · needs ~752.09 GB, NVIDIA only · from huggingface.co/LGAI-EXAONE
vllm serve LGAI-EXAONE/K-EXAONE-2.0-750B-A37B-FP8 --served-model-name K-EXAONE-2.0-750B-A37B-FP8 --trust-remote-code --tensor-parallel-size 8 --kv-cache-dtype fp8_e4m3 --max-num-seqs 256 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_xml

Weights and documentation: huggingface.co/LGAI-EXAONE ↗

$ how the score was worked out · LGAI-EXAONE/K-EXAONE-2.0-750B… 99%

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	licence
LGAI-EXAONE/K-EXAONE-2.0-750B-A37B model card · lab-reported	11	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%	Apache-2.0
prior · pull toward the middle		▇ 1%
weighted average		= 67.2

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-10-11 02:20 UTC · sources and licences · json