{ "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-11-11", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "tool": { "slug": "exaone-4-5-33b", "name": "EXAONE 4.5 33B", "vendor": "LG AI Research", "vendorUrl": "https://www.lgresearch.ai", "kind": "weights", "category": "open-weights", "summary": "EXAONE 4.5 33B is LG AI Research's open-weight vision-language model, 34B parameters, with tool calling, switchable reasoning and a 256K context, strong on documents and Korean. Its licence allows research and education only, with LG's own GGUF from 18 GB.", "url": "https://www.anchorterminal.com/tools/exaone-4-5-33b", "markdownUrl": "https://www.anchorterminal.com/tools/exaone-4-5-33b.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/exaone-4-5-33b.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/exaone-4-5-33b.json", "dns": "exaone-4-5-33b.grade.anchorterminal.com", "repo": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B", "license": "EXAONE AI Model License Agreement 1.2 - NC (research and education only)", "transports": [], "packages": [ { "registry": "huggingface", "name": "LGAI-EXAONE/EXAONE-4.5-33B" }, { "registry": "huggingface", "name": "LGAI-EXAONE/EXAONE-4.5-33B-GGUF" }, { "registry": "huggingface", "name": "LGAI-EXAONE/EXAONE-4.5-33B-AWQ" }, { "registry": "huggingface", "name": "LGAI-EXAONE/EXAONE-4.5-33B-FP8" } ], "auth": "none", "authNotes": "The weights download from Hugging Face with no account or token. None of the four repositories is gated, though the licence says downloading is agreeing to it (checked 2026-11-10).", "pricing": "free", "pricingNotes": "Free to download for research and education. The EXAONE AI Model License Agreement 1.2 - NC prohibits commercial use of the model, its derivatives and its output without a separate agreement with LG. Running it yourself takes 20.6 GB at IQ4_XS with its projector, up to 68.7 GB in BF16 (checked 2026-11-10).", "priceSummary": "Free", "where": "local", "x402": { "level": "no", "evidence": "Weights you download and run yourself, so there is nothing to pay per call and no payment protocol (checked 2026-11-10).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-10" }, "docsUrl": "https://github.com/LG-AI-EXAONE/EXAONE-4.5", "capabilities": [ "model.open-weights" ], "tags": [ "open-weights", "model", "multimodal", "vision", "documents", "korean", "reasoning", "tool-calling", "long-context", "non-commercial", "safetensors", "gguf", "awq", "fp8", "self-hosted" ], "lastRelease": "2026-07-06", "graded": true, "anchor": { "graded": true, "score": 61, "grade": "C", "agentReady": false, "rank": 0, "ranked": false, "notRankedWhy": "An open-weight model, graded by the open-weight checklist and ranked among open-weight models, not against tools", "rankOf": 1518, "categoryRank": 51, "methodology": "0.7", "run": "2026-10-01", "scores": { "interface": 85, "reach": 65, "risk": 50, "upkeep": 50 }, "pending": [ "performance", "tasks" ], "breakdown": [ { "key": "capability", "name": "Capability", "weight": 35, "effectiveWeight": 0, "pending": true, "points": 0, "reason": "Pending for this listing. Nobody independent has published results for it in at least two areas yet, so its weight is shared across the other four and the total is capped at 61 until they do or our own runs start." }, { "key": "reach", "name": "Reach", "weight": 25, "effectiveWeight": 38.46, "score": 65, "points": 25, "reason": "Memory 20, the smallest build at 4 bits is LG's IQ4_XS GGUF, 18.03 GB with the 2.57 GB projector, and 20.60 GB × 1.2 = 24.7 GB, within 32 GB. Without the projector, for text only, it would fit 24 GB. Runtimes 15, Released vLLM 0.33.0 has exaone4_5.py, Transformers 5.22.2 has the exaone4_5 model, and llama.cpp b12242 has the architecture, with LG's own GGUF. Three at 5 each. Builds 15, LG publishes GGUF at five quantised precisions, an AWQ build and an FP8 build. Terms 5, Research and educational use only. Download 10, the Hub API reports gated false for all four repositories on 10 November 2026, so the files download with no account and no click-through, though the licence says downloading is acceptance. Hosted 0, Hugging Face maps no inference provider to it, and the card names no API that serves it. 20 + 15 + 15 + 5 + 10 + 0 = 65.", "checked": "2026-11-10" }, { "key": "risk", "name": "Risk", "weight": 20, "effectiveWeight": 30.77, "score": 50, "points": 15.38, "reason": "File safety 25, safetensors and GGUF only, no .bin, .pt or .pth files and no Python in the repository. Provenance 15, the files sit in the LGAI-EXAONE organisation (LG AI Research on the Hub), which the card's GitHub repository shares a name with and the licence in the repository is LG's (10). Safety evaluation 0, no safety results on the card. Agent guidance 5, General cautions: the limitation section lists inappropriate, biased, incorrect and outdated answers and tells users not to try to induce harmful output. Licence risk 5, 15 less 5 for field-of-use limits beyond the law (no commercial use, no building competing models) and 5 for terms the licensor can change or revoke (LG may amend the agreement at any time, and the licence is revocable). Security programme 0, no way to report a flaw on the card beyond a support contact, and www.lgresearch.ai has no security.txt. 25 + 15 + 0 + 5 + 5 + 0 = 50.", "checked": "2026-11-10" }, { "key": "interface", "name": "Agent interface", "weight": 10, "effectiveWeight": 15.38, "score": 85, "points": 13.08, "reason": "Chat template 20, chat_template.jinja in the repository (5,930 characters). Tool-calling format 25, the card's vLLM command names the parser, hermes with auto tool choice, and its agentic example calls tools from MCP servers through the OpenAI-compatible API, so declaration and return follow that API. Trained tool calling 20, the card reports agentic tool use among its language results and shows an agent run with several tool servers, and the template renders every call in a turn. Structured output 0, Nothing on the card about JSON or schema-constrained output. Reasoning control 10, enable_thinking is on by default and the card says to set it false for the non-reasoning mode, with an example of each. Context 10, 262,144 tokens per the card's configuration table and config.json, and the card's serve commands run at that length. 20 + 25 + 20 + 0 + 10 + 10 = 85.", "checked": "2026-11-10" }, { "key": "upkeep", "name": "Documentation and upkeep", "weight": 10, "effectiveWeight": 15.38, "score": 50, "points": 7.69, "reason": "Model card 20, Uses (the quickstart's question answering, document and agent examples), limitations, the architecture, the parameters, the context, the languages (in the metadata) and a knowledge cut-off of December 2024 are all on the card. Training data 5, A knowledge cut-off and one line on excluding personal, harmful and biased content, with no sources or sizes. Last change 10, the main repository last changed on 6 July 2026 (dependencies moved to official releases), 127 days before this check, and the FP8 repository on 9 August, 93 days. Responsiveness 0, none of the 5 most recent discussions that aren't pull requests (13 April to 4 May 2026) has a reply from an LGAI-EXAONE member, counted on 10 November by author flags and dates with a script, text not read. Version policy 5, EXAONE 4.0 and 4.0.1 stay in their own repositories and 4.5 came out in new ones, with no dated note on the older cards that we read. File health 10, Config, template and card agree on the architecture and the 262,144 tokens, and a July commit moved the dependencies to released versions. 20 + 5 + 10 + 0 + 5 + 10 = 50.", "checked": "2026-11-10" } ], "assessment": { "date": "2026-11-10", "basis": "public evidence", "confidence": "low", "notes": { "capability": "Pending for this listing. Independent results were found in 0 areas (none), and Capability needs 2. Its weight is shared across the other four categories and the total is capped at 61 until there are more, or our own runs start.", "interface": "Chat template 20, chat_template.jinja in the repository (5,930 characters). Tool-calling format 25, the card's vLLM command names the parser, hermes with auto tool choice, and its agentic example calls tools from MCP servers through the OpenAI-compatible API, so declaration and return follow that API. Trained tool calling 20, the card reports agentic tool use among its language results and shows an agent run with several tool servers, and the template renders every call in a turn. Structured output 0, Nothing on the card about JSON or schema-constrained output. Reasoning control 10, enable_thinking is on by default and the card says to set it false for the non-reasoning mode, with an example of each. Context 10, 262,144 tokens per the card's configuration table and config.json, and the card's serve commands run at that length. 20 + 25 + 20 + 0 + 10 + 10 = 85.", "reach": "Memory 20, the smallest build at 4 bits is LG's IQ4_XS GGUF, 18.03 GB with the 2.57 GB projector, and 20.60 GB × 1.2 = 24.7 GB, within 32 GB. Without the projector, for text only, it would fit 24 GB. Runtimes 15, Released vLLM 0.33.0 has exaone4_5.py, Transformers 5.22.2 has the exaone4_5 model, and llama.cpp b12242 has the architecture, with LG's own GGUF. Three at 5 each. Builds 15, LG publishes GGUF at five quantised precisions, an AWQ build and an FP8 build. Terms 5, Research and educational use only. Download 10, the Hub API reports gated false for all four repositories on 10 November 2026, so the files download with no account and no click-through, though the licence says downloading is acceptance. Hosted 0, Hugging Face maps no inference provider to it, and the card names no API that serves it. 20 + 15 + 15 + 5 + 10 + 0 = 65.", "risk": "File safety 25, safetensors and GGUF only, no .bin, .pt or .pth files and no Python in the repository. Provenance 15, the files sit in the LGAI-EXAONE organisation (LG AI Research on the Hub), which the card's GitHub repository shares a name with and the licence in the repository is LG's (10). Safety evaluation 0, no safety results on the card. Agent guidance 5, General cautions: the limitation section lists inappropriate, biased, incorrect and outdated answers and tells users not to try to induce harmful output. Licence risk 5, 15 less 5 for field-of-use limits beyond the law (no commercial use, no building competing models) and 5 for terms the licensor can change or revoke (LG may amend the agreement at any time, and the licence is revocable). Security programme 0, no way to report a flaw on the card beyond a support contact, and www.lgresearch.ai has no security.txt. 25 + 15 + 0 + 5 + 5 + 0 = 50.", "upkeep": "Model card 20, Uses (the quickstart's question answering, document and agent examples), limitations, the architecture, the parameters, the context, the languages (in the metadata) and a knowledge cut-off of December 2024 are all on the card. Training data 5, A knowledge cut-off and one line on excluding personal, harmful and biased content, with no sources or sizes. Last change 10, the main repository last changed on 6 July 2026 (dependencies moved to official releases), 127 days before this check, and the FP8 repository on 9 August, 93 days. Responsiveness 0, none of the 5 most recent discussions that aren't pull requests (13 April to 4 May 2026) has a reply from an LGAI-EXAONE member, counted on 10 November by author flags and dates with a script, text not read. Version policy 5, EXAONE 4.0 and 4.0.1 stay in their own repositories and 4.5 came out in new ones, with no dated note on the older cards that we read. File health 10, Config, template and card agree on the architecture and the 262,144 tokens, and a July commit moved the dependencies to released versions. 20 + 5 + 10 + 0 + 5 + 10 = 50." }, "sources": [ { "what": "model card", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B", "seen": "2026-11-10" }, { "what": "gating, dates, parameter count, downloads, likes, file list and hashes", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B?blobs=true", "seen": "2026-11-10" }, { "what": "LG's GGUF repository: files and sizes", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B-GGUF?blobs=true", "seen": "2026-11-10" }, { "what": "the AWQ repository", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B-AWQ?blobs=true", "seen": "2026-11-10" }, { "what": "the FP8 repository", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B-FP8?blobs=true", "seen": "2026-11-10" }, { "what": "inference providers, none mapped", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B?expand[]=inferenceProviderMapping", "seen": "2026-11-10" }, { "what": "commit history", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B/commits/main", "seen": "2026-11-10" }, { "what": "architecture and positions", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/raw/main/config.json", "seen": "2026-11-10" }, { "what": "chat template", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/raw/main/chat_template.jinja", "seen": "2026-11-10" }, { "what": "the licence", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/raw/main/LICENSE", "seen": "2026-11-10" }, { "what": "discussions, replies counted by script from authors, dates and member flags", "url": "https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B/discussions", "seen": "2026-11-10" }, { "what": "the organisation's profile on the Hub, not marked verified", "url": "https://huggingface.co/api/organizations/LGAI-EXAONE/overview", "seen": "2026-11-10" }, { "what": "the organisation's other repositories", "url": "https://huggingface.co/api/models?author=LGAI-EXAONE", "seen": "2026-11-10" }, { "what": "robots.txt for www.lgresearch.ai", "url": "https://www.lgresearch.ai/robots.txt", "seen": "2026-11-10" }, { "what": "LG AI Research's home page", "url": "https://www.lgresearch.ai/", "seen": "2026-11-10" }, { "what": "security.txt, 404", "url": "https://www.lgresearch.ai/.well-known/security.txt", "seen": "2026-11-10" }, { "what": "vLLM v0.33.0, exaone4_5.py", "url": "https://raw.githubusercontent.com/vllm-project/vllm/v0.33.0/vllm/model_executor/models/exaone4_5.py", "seen": "2026-11-10" }, { "what": "Transformers v5.22.2 exaone4_5", "url": "https://github.com/huggingface/transformers/tree/v5.22.2/src/transformers/models/exaone4_5", "seen": "2026-11-10" }, { "what": "llama.cpp b12242, the exaone architecture (shallow clone)", "url": "https://raw.githubusercontent.com/ggml-org/llama.cpp/b12242/src/llama-arch.cpp", "seen": "2026-11-10" } ], "openQuestions": [ "It's a non-commercial model in a directory mostly used for commercial work. The terms are stated first in the notes for agents", "Memory reads the IQ4_XS file with its projector, 24.7 GB with the fifth added, which just misses the 24 GB band", "Provenance takes the LGAI-EXAONE organisation as LG's on the licence and the card. LG's site gave too little text to confirm a link", "Security programme is 0 on the card and the research site. LG's corporate security pages weren't looked up", "Capability is pending because no independent results were looked up on the day", "unchecked on 10 November: the technical report (arXiv 2604.08644), the blog post, the GitHub repository, TensorRT-LLM's docs, Ollama's library, LM Studio's catalogue, and LG's corporate pages" ] }, "negative": 0, "verdict": "A well-documented vision-language model with a full set of builds, tool calling, switchable reasoning, a 256K context and a stated knowledge cut-off. The licence decides whether you can use it: research and education only, output included, and LG can change the terms. Nothing answered on the repository and no safety results.", "bestFor": "Researchers and students working on documents, Korean or multimodal agents who don't need commercial rights", "strengths": [ "A complete card, down to a knowledge cut-off", "GGUF, AWQ and FP8 builds from LG", "Released vLLM, Transformers and llama.cpp run it with no repository code", "Tool calling and reasoning control documented with parsers", "A 256K context, served on one large GPU" ], "weaknesses": [ "Research and education only. No commercial use of the model or its output", "LG can amend the licence at any time, and it's revocable", "No building models that compete with LG's", "No reply from LG on the repository's discussions", "Not changed since the summer", "No safety results or training data sources on the card" ], "agentNotes": [ "Don't use it, or anything it writes, commercially. The licence covers the output too", "Serve it with vLLM and `--reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser hermes`", "Set `enable_thinking` false for a direct answer. Thinking is on by default", "Use temperature 1.0, top_p 0.95 and presence_penalty 1.5 in general, and temperature 0.6 with top_k 20 for documents and Korean", "For llama.cpp, load a projector file from LG's GGUF repository with the weights for image input", "Its knowledge stops at December 2024" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "low", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 61 } ], "editorialScores": { "interface": 85, "reach": 65, "risk": 50, "upkeep": 50 }, "provenanceScore": 49, "checked": { "categories": { "interface": "2026-11-10", "reach": "2026-11-10", "risk": "2026-11-10", "upkeep": "2026-11-10" }, "last": "2026-11-10", "oldest": "2026-11-10", "stale": false, "staleAfterDays": 90 }, "scoreHistory": [ { "date": "2026-11-10", "grade": "C", "kind": "listed", "method": "0.7", "score": 61 } ] }, "connect": { "install": "pip install \"vllm\u003e=0.33.0\" \"transformers\u003e=5.8.0\"\nvllm serve LGAI-EXAONE/EXAONE-4.5-33B --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser hermes", "http": "# vLLM serves an OpenAI-compatible API at http://localhost:8000/v1. The card recommends temperature 1.0,\n# top_p 0.95 and presence_penalty 1.5, and enable_thinking false for a direct answer" }, "letme": { "capability": "https://letme.dev/model.open-weights", "tool": "https://letme.dev/exaone-4-5-33b" }, "weights": { "trainedBy": "LG AI Research", "params": "34B", "context": 262144, "goodFor": "Research on document understanding, Korean and agent work with images, where non-commercial terms are acceptable.", "runtimes": [ "llama.cpp", "vLLM", "Transformers" ], "builds": [ { "name": "EXAONE-4.5-33B-GGUF IQ4_XS", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF/blob/main/EXAONE-4.5-33B-IQ4_XS.gguf", "who": "publisher", "precision": "iq4_xs", "sizeGB": 20.6, "memoryGB": 32, "format": "gguf", "runsOn": [ "llama.cpp" ], "note": "18.03 GB with the 2.57 GB F16 projector" }, { "name": "EXAONE-4.5-33B-GGUF Q4_K_M", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF/blob/main/EXAONE-4.5-33B-Q4_K_M.gguf", "who": "publisher", "precision": "q4_k_m", "sizeGB": 22.62, "memoryGB": 32, "format": "gguf", "runsOn": [ "llama.cpp" ], "note": "20.05 GB with the 2.57 GB F16 projector" }, { "name": "EXAONE-4.5-33B-AWQ", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-AWQ", "who": "publisher", "precision": "awq", "sizeGB": 22.75, "memoryGB": 32, "format": "safetensors", "runsOn": [ "vLLM" ], "note": "LG's 4-bit AWQ build" }, { "name": "EXAONE-4.5-33B-GGUF Q8_0", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF/blob/main/EXAONE-4.5-33B-Q8_0.gguf", "who": "publisher", "precision": "q8_0", "sizeGB": 37.71, "memoryGB": 48, "format": "gguf", "runsOn": [ "llama.cpp" ], "note": "35.14 GB with the 2.57 GB F16 projector" }, { "name": "EXAONE-4.5-33B-FP8", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-FP8", "who": "publisher", "precision": "fp8", "sizeGB": 41.31, "memoryGB": 64, "format": "safetensors", "runsOn": [ "vLLM" ], "note": "LG's FP8 build" }, { "name": "EXAONE-4.5-33B (BF16 safetensors)", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B", "who": "publisher", "precision": "bf16", "sizeGB": 68.7, "memoryGB": 96, "format": "safetensors", "runsOn": [ "vLLM", "Transformers" ], "note": "Two shards" } ], "run": "llama-server -hf LGAI-EXAONE/EXAONE-4.5-33B-GGUF:Q4_K_M --jinja", "runBuild": "EXAONE-4.5-33B-GGUF Q4_K_M", "checked": "2026-11-10", "minMemoryGB": 32 }, "family": { "lab": "LG AI Research", "family": "EXAONE 4", "url": "https://www.anchorterminal.com/open-weights/#lg-ai-research", "others": [ { "name": "EXAONE-4.0.1-32B", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.0.1-32B" }, { "name": "EXAONE-4.0-1.2B", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.0-1.2B" } ] }, "notable": [ "23,671 downloads in the last 30 days and 193 likes on Hugging Face on 10 November 2026 (https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B)", "The licence allows research and educational use only, bars commercial use of the output as well as the model, and bars using it to build models that compete with LG's (https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/blob/main/LICENSE)", "LG can modify the licence at any time by posting a new version, and continued use counts as acceptance (https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/blob/main/LICENSE)", "The card gives a knowledge cut-off of December 2024, which few open-weight cards do (https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B)", "LG publishes GGUF builds from IQ4_XS to BF16 with projector files, and AWQ and FP8 builds (https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF)", "The card says one H200, or four A100 40 GB cards, serves it at the full 256K context (https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B)", "None of the 5 discussions on the repository has a reply from an LG member (https://huggingface.co/api/models/LGAI-EXAONE/EXAONE-4.5-33B/discussions)" ], "area": "models", "details": [ { "label": "Surface graded", "value": "The open weights at huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B and the GGUF, AWQ and FP8 repositories beside it, read from the model card, the files and their metadata, the config, the chat template, the LICENSE, the commit history and the runtimes' own code at their latest releases on 10 November 2026. LG AI Research's site was read for who it is and gave little text. The technical report, the blog post and the GitHub repository weren't read. Nothing was downloaded or run" }, { "label": "Model", "value": "34,350,097,664 parameters in the files. The card gives 31.7B for the language model and 1.29B for the vision encoder: 64 layers that interleave sliding-window and global attention, with one extra layer for multi-token prediction. LG's first open-weight vision-language model, built on the EXAONE 4.0 framework" }, { "label": "Files", "value": "Two safetensors shards in BF16 (68.7 GB), chat_template.jinja, config, processor and tokenizer files and the LICENSE. No pickle files and no Python. The GGUF repository has six precisions and three projector files. Hugging Face lists a SHA-256 for the large files" }, { "label": "Licence", "value": "EXAONE AI Model License Agreement 1.2 - NC, from LG Management Development Institute Co., Ltd. A revocable licence for research and educational use. Commercial use of the model, derivatives or output needs a separate agreement, and so does using it to develop models that compete with LG's. LG may amend the agreement at any time. Korean law governs" }, { "label": "Runtimes", "value": "The card documents TensorRT-LLM, vLLM, SGLang from a fork, and Transformers 5.8.0 or later, and says llama.cpp supports it. Released vLLM 0.33.0 and Transformers 5.22.2 have the exaone4_5 model, and llama.cpp b12242 has the architecture, with LG's own GGUF. SGLang 0.5.23 has no code for it" }, { "label": "Context", "value": "262,144 tokens, per the card's configuration table and config.json" }, { "label": "Hardware", "value": "One H200, or four A100 40 GB cards with tensor parallelism, for the full 256K context, per the card. IQ4_XS with the projector is 20.6 GB" }, { "label": "Training data", "value": "Not described on the card beyond a knowledge cut-off of December 2024 and a line that LG tries to exclude personal, harmful and biased content. The technical report wasn't read" }, { "label": "Languages", "value": "English, Korean, Spanish, German, Japanese and Vietnamese, per the card's metadata" }, { "label": "Tool calling", "value": "The card's vLLM command sets auto tool choice with the hermes tool-call parser, and its agentic example drives the model through LangChain with MCP servers. Its results table has a row group for agentic tool use" }, { "label": "Reasoning", "value": "enable_thinking is on by default, and the card says to set it false for the non-reasoning mode. The qwen3 reasoning parser returns the reasoning apart from the answer" }, { "label": "Recommended settings", "value": "temperature 1.0, top_p 0.95 and presence_penalty 1.5 in general; temperature 0.6 with top_k 20 for OCR, documents and Korean; no presence penalty for text-only input. The card says the model prefers to answer in a boxed format" } ], "provenance": { "legalEntity": "LG Management Development Institute Co., Ltd., the licensor named in the model's licence. The card names LG AI Research", "domain": "lgresearch.ai", "domainRegistered": "", "endpointOnVendorDomain": null, "terms": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/blob/main/LICENSE", "privacy": "", "statusPage": "", "changelog": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/commits/main", "securityTxt": "none", "checked": "2026-11-10", "notes": [ "The card and the Hub organisation name LG AI Research, and the licence names LG Management Development Institute Co., Ltd. as licensor. How the two relate wasn't read first-hand.", "www.lgresearch.ai's robots.txt allows reading. Its home page gave little text and its about page a 404, so no people or ownership facts were taken from it.", "The domain's registration date wasn't looked up. www.lgresearch.ai has no security.txt (404 on 10 November 2026).", "Nothing is hosted under this listing, so the model licence stands in for terms of service and no privacy policy is scored.", "The changelog is the commit history, 8 commits from 9 April to 6 July 2026." ], "score": 49, "checks": [ { "check": "Legal entity named", "value": "LG Management Development Institute Co., Ltd., the licensor named in the model's licence. The card names LG AI Research", "points": 20, "max": 20, "state": "ok" }, { "check": "Domain age", "value": "lgresearch.ai, no registry record we could read", "points": 0, "max": 15, "state": "no" }, { "check": "Endpoint on the vendor's domain", "value": "no hosted endpoint", "points": 0, "max": 0, "state": "na" }, { "check": "Terms of service", "value": "published, not read by us yet", "points": 4, "max": 10, "state": "part" }, { "check": "Privacy policy", "value": "nothing hosted, not scored", "points": 0, "max": 0, "state": "na" }, { "check": "Status page", "value": "not found", "points": 0, "max": 10, "state": "no" }, { "check": "Changelog", "value": "published", "points": 10, "max": 10, "state": "ok" }, { "check": "security.txt", "value": "not found", "points": 0, "max": 10, "state": "no" }, { "check": "People named", "value": "not researched yet", "points": 0, "max": 0, "state": "na" }, { "check": "Support route", "value": "help pages, a community", "points": 5, "max": 5, "state": "ok" } ] }, "humans": { "slug": "exaone-4-5-33b", "checked": "2026-11-10", "company": "LG AI Research", "peopleNote": "The model card credits LG AI Research and names no people, and the licence names LG Management Development Institute Co., Ltd. as licensor. LG AI Research's site, read on 10 November 2026, gave a menu and little other text, and its about page answered 404, so no people are listed and ownership is recorded as unknown.", "companySocial": { "github": "https://github.com/LG-AI-EXAONE" }, "support": { "urls": [ "https://github.com/LG-AI-EXAONE/EXAONE-4.5" ], "community": [ { "label": "Hugging Face discussions", "url": "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/discussions", "kind": "forum" } ] }, "sources": [ "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B", "https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B/blob/main/LICENSE", "https://www.lgresearch.ai/" ], "ownership": { "status": "unknown", "checked": "2026-11-10" } }, "pageJsonUrl": "https://www.anchorterminal.com/tools/exaone-4-5-33b.json" } }