LA
LG AI Research
Language model · South Korea ·Jul 2026

K-EXAONE 2.0

Language Open Source
750B
Parameters
262.1K
Context

About K-EXAONE 2.0

K-EXAONE 2.0 is an AI model developed by LG AI Research, in the language category, released in 2026, made available as an open-source model with 750B params. Korea's largest AI foundation model with 750 billion parameters, released under the Apache 2.0 license to Hugging Face. It supports 10 languages and demonstrates strong performance in coding, agentic tool-use, long-context comprehension, and instruction-following benchmarks.

On this page you'll find K-EXAONE 2.0's full specifications, including a context window of 262.1K tokens. Review provider pricing and benchmark scores below, or compare K-EXAONE 2.0 head-to-head with other language models.

→ More models by LG AI Research → All language models → AI benchmarks leaderboard

Links

Announcement ↗ Hugging Face ↗

Run K-EXAONE 2.0 locally — what it takes

K-EXAONE 2.0 (750B) is too large to run on a single consumer or Apple machine. At Q4 its weights alone are about 420 GB — it needs a multi-GPU / datacenter setup:

  • ≈ 2× AMD Instinct MI325X
  • ≈ 3× NVIDIA B200

Lower quantization (Q3/Q2) or CPU + system-RAM offload (e.g. ktransformers) reduces the requirement at the cost of speed.

Software support

·ExLlamaV2 ·KoboldCpp ·LM Studio ·MLX ·Ollama ·SGLang ·TensorRT-LLM ·llama.cpp ·vLLM ?KTransformers ?LM Studio Bionic ?Open WebUI ?text-generation-webui

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Community

Would you run K-EXAONE 2.0 again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

More models like K-EXAONE 2.0

AI
DMind-1
 AI
COLLECTIVE-1
 AI
Cogito 1 series
 AI
Euler (Euler大模型)
· 2.5B
 AI
SmolLM-1.7B
· 1.7B
 AI
Llama-3-70B-Synthia-v3.5
· 70B

Frequently asked questions

What is K-EXAONE 2.0? ▶

Korea's largest AI foundation model with 750 billion parameters, released under the Apache 2.0 license to Hugging Face. It supports 10 languages and demonstrates strong performance in coding, agentic tool-use, long-context comprehension, and instruction-following benchmarks. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created K-EXAONE 2.0? ▶

K-EXAONE 2.0 was developed by LG AI Research and released in 2026.

Is K-EXAONE 2.0 open source or proprietary? ▶

K-EXAONE 2.0 is released as open source — its weights are publicly available to download, run locally and fine-tune.

What is K-EXAONE 2.0's context window? ▶

K-EXAONE 2.0 supports a context window of 262.1K tokens, and has 750B params.

How much does K-EXAONE 2.0 cost? ▶

Pricing for K-EXAONE 2.0 depends on the provider. See the providers table on this page for the latest API rates.

How does K-EXAONE 2.0 perform on benchmarks? ▶

Benchmark scores for K-EXAONE 2.0 are listed on this page as they are published. You can compare it head-to-head with other models on the GenAIList compare tool.

Let's Run K-EXAONE 236B - Korea's LARGEST Local AI answer to QWEN | Developer Review

Know a great video about K-EXAONE 2.0?