[whatmodel.sh](../../README.md) / GLM-5.3-Flash # GLM-5.3-Flash Z.ai · released 2026-08 · 320B params (18B active) · mit **score 72.4** ±0.7 · #56 of 732 · from 21 benchmark results across 8 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks GLM-5.3-Flash among every model on that task or test, worst to best.}

task                vs the others         rank  counts  tests
agent tasks         ▁▁▃▆█▄▅▅▄●▂▁    #43 of 313     23%      7
coding              ▁▁▁▃▅▇█▇▅▄●▁    #29 of 469     13%      3
chat & writing      ▁▁▁▃▆▄▆█▆▇●▃   #27 of 402*    4.7%      1
math                ▁▁▃▅█▆▄█●▂▁▁    #52 of 431     39%      3
vision & documents  ▁▁▂▂▁▃▄▂▆▇●▄   #30 of 153*    3.2%      2
web search          ▁▁▁▅▄●▃▄·▁·▁    #39 of 84*    1.3%      1

agent tasks · #43 of 313 · 23% of its score

Beats [GPT-5.6 Luna](gpt-5-6-luna.md), [Gemini 3.7 Flash](gemini-3-7-flash.md), [Muse Spark Open](muse-spark-open.md) here, though they rank higher overall. Trails [GLM-5.2](glm-5-2.md) here, though they rank lower overall.

test                                     result  vs the others in its run  difficulty            counts  why
Arena Agent (October 2026 aggrega…  -0.0233 IPS  ▁▁▁▃▃●█▁▂▂▁▁ beats 21/59  avg -0.379 IPS           12%  sharp
Terminal-Bench 4.0                        35.8%  █▅▅▅▅·●▃▃▅▅▅ beats 11/20  very hard · avg 0.5%    3.7%  sharp
Terminal-Bench 2.1 (maker)                84.3%  ▁·▁·▁▁▄▃▆█▅● beats 51/63  hard · avg 25%          2.2%  sharp · small run
Video2World V2WScore (coding agen…        25.7%  ●··▄█·▄····▄ beats 0/6    very hard · avg 7.3%    1.8%  at its level
Humanity's Last Exam with tools (…        55.3%  ▁·▁▂▂▁▄▄▇●▄▅ beats 45/62  hard · avg 16%          1.3%  at its level · small run
Agents' Last Exam (metric/harness…        26.3%  ▂·▂·●▄▂·▂▂▄▄ beats 3/14   very hard · avg 7.0%    0.9%  at its level · small run
DeepSWE v1.1 (resolved) (maker)           63.4%  ▂····▂▁▁▆●██ beats 11/31  hard · avg 20%          0.9%  at its level · small run

coding · #29 of 469 · 13% of its score

Beats [GPT-5.5](gpt-5-5.md), [GPT-5.6 Terra](gpt-5-6-terra.md), [Claude Opus 4.8](claude-opus-4-8.md) here, though they rank higher overall.

test                                result  vs the others in its run    difficulty      counts  why
Arena WebDev (Code Arena)           1606.6  ▁▁▃▂▅█▃▆●▅▁▂ beats 101/129  avg 981           8.4%  sharp
Luau Leetcode fully solved (Torpe…   68.8%  ▁▁▁▁·▁▂▃●▃█▇ beats 22/49    hard · avg 17%    3.6%  sharp · at its level
DeepSWE v1.1 (resolved) (maker)      63.4%  ▂····▂▁▁▆●██ beats 11/31    hard · avg 20%    0.9%  at its level · small run

chat & writing · #27 of 402 · 4.7% of its score

Beats [GPT-6 Astra](gpt-6-astra.md), [GPT-6 Sol](gpt-6-sol.md), [GPT-5.6 Terra](gpt-5-6-terra.md) here, though they rank higher overall. Trails [Gemini 3.6 Flash](gemini-3-6-flash.md) here, though they rank lower overall.

test            result  vs the others in its run    difficulty  counts  why
LMArena (text)  1476.5  ▁▁▁▂▃▃▃▆▆▇●▃ beats 357/385  avg 1260      4.7%  blunt

math · #52 of 431 · 39% of its score

Beats [Grok 4.7](grok-4-7.md) here, though they rank higher overall. Trails [A.X K2 (688B-A33B)](ax-k2.md), [DeepSeek V4 Flash (Preview, Apr 2026)](deepseek-v4-flash-preview.md), [GPT-5](gpt-5.md) here, though they rank lower overall.

test                       result  vs the others in its run    difficulty            counts  why
FrontierMath tiers 1–3 v2   55.8%  ▂··▁▁▃▄▄●▆▄▃ beats 41/86    very hard · avg 1.2%     22%  sharp · at its level
FrontierMath tier 4 v2      17.1%  ▁▂▁▂●▅▂▃▂▁▁▁ beats 20/67    very hard · avg 0.1%     16%  sharp · at its level
OTIS mock AIME 2024–2025    93.9%  ▃▃▄▃▃▃▇▇█●▂▅ beats 157/197  very hard · avg 7.2%    0.9%  sharp · maxed out · noisy

vision & documents · #30 of 153 · 3.2% of its score

Beats [Claude Sonnet 5.5](claude-sonnet-5-5.md), [GPT-6 Sol](gpt-6-sol.md), [MiMo-V2.6-Pro](mimo-v2-6-pro.md) here, though they rank higher overall. Trails [Qwen3.6 Plus](qwen3-6-plus.md), [Qwen3.7 Plus](qwen3-7-plus.md), [Gemini 3 Flash (Preview)](gemini-3-flash-preview.md) here, though they rank lower overall.

test                                result  vs the others in its run    difficulty        counts  why
Arena vision (style controlled)     1271.2  ▁▁▂▂▂▂▃▄▅▅●▇ beats 124/152  avg 1061            3.1%  blunt
PureDocBench document parsing (Av…   79.7%  ▂·▁·▂▂▅█·▁▂● beats 22/22    medium · avg 62%    0.1%  blunt · at its level

web search · #39 of 84 · 1.3% of its score

Beats [Gemini 3.1 Pro (Preview)](gemini-3-1-pro.md), [Gemini 3.5 Flash](gemini-3-5-flash.md), [Qwen3.7 Max](qwen3-7-max.md) here, though they rank higher overall. Trails [GPT-4o mini (2024-07-18)](gpt-4o-mini-2024-07-18.md), [Mistral Small 3.1 24B (2503)](mistral-small-3-1.md), [Llama 4 Scout (109B-A17B)](llama-4-scout.md) here, though they rank lower overall.

test                                result  vs the others in its run  difficulty      counts  why
Humanity's Last Exam with tools (…   55.3%  ▁·▁▂▂▁▄▄▇●▄▅ beats 45/62  hard · avg 16%    1.3%  at its level · small run

other tests (6)

test                                result  vs the others in its run    difficulty            counts  why
Mystery Game Puzzles (Epoch)          8.0%  ●··▂▃▇▄█▂▄▂▁ beats 15/77    very hard · avg 9.4%    5.8%  sharp · noisy
GPQA Diamond                         90.2%  ▁▁▁▁▂▄▆▄▅▇●▅ beats 175/211  hard · avg 39%          3.9%  sharp
Chess Puzzles (Epoch)                14.0%  █▁▁▁▁●▃▃▃▂▁▁ beats 68/137   very hard · avg 5.8%    3.8%  sharp · noisy
Icelandic LLM Leaderboard average…   27.9%  ▁▁▃▂▂●▁▃▄▄▄█ beats 11/38    very hard · avg 14%     1.8%  at its level
RobloxQA v2.0 multiple choice (To…   94.6%  ▁▁▃▁█▄●▃▆▂▂▂ beats 28/49    easy · avg 77%          1.3%  maxed out
AutomationBench v1.0.6 (maker)       48.8%  ▃···▃·▃▃··●▅ beats 4/8      very hard · avg 7.4%    1.0%  at its level · small run

## 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
Epoch AI                                        6   53% ▇▇▇▇▇▇
LMArena leaderboard dataset                     4   28% ▇▇▇
zai-org/GLM-5.3-Flash model card (submi…        5    6% ▇
Roblox LLM Leaderboard                          2    5% ▇
Terminal-Bench 4.0 official leaderboard…        1    4% ▇
Video2World: Benchmarking Coding Agents…        1    2% ▇
Miðeind Icelandic LLM Leaderboard               1    2% ▇
How Far Is Document Parsing from Solved…        1    0% ▇
prior: pull toward the middle                        0% ▇
weighted average                                   = 72.4

_{Licences: Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · zai-org/GLM-5.3-Flash model card ([MIT](../../licenses/zai-org--GLM-5.3-Flash.txt)) · Roblox LLM Leaderboard ([MIT](../../licenses/roblox-llm-leaderboard-MIT.txt)) · Terminal-Bench 4.0 official leaderboard submissions ([Apache-2.0](../../licenses/terminal-bench-leaderboard-submissions-Apache-2.0.txt)) · Video2World: Benchmarking Coding Agents for Interactive World Modeling from Embodied Videos ([CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/)) · Miðeind Icelandic LLM Leaderboard ([CC BY 4.0](../../licenses/mideind-icelandic-llm-leaderboard-CC-BY-4.0.txt)) · How Far Is Document Parsing from Solved? PureDocBench: A Source-Traceable Benchmark across Clean ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))} ## How to run it ### Through an API api (Z.AI) · $0.24/M blended · $0.15 in / $0.5 out per 1M tokens · [docs.z.ai](https://docs.z.ai/guides/overview/pricing) _{[price source](https://github.com/anomalyco/models.dev/blob/a668908cfe7073faa0363699486fc41a3ff52412/providers/zai/models/glm-5.3-flash.toml): Standard base token rate from models.dev. Provider-specific tiers, long-context surcharges, caching and promotions may differ. · [context source](https://github.com/anomalyco/models.dev/blob/a668908cfe7073faa0363699486fc41a3ff52412/models/zhipuai/glm-5.3-flash.toml)} ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run GLM-5.3-Flash (ud-q4_k_xl, needs ~192.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF/tree/main)) ```sh llama cli -hf unsloth/GLM-5.3-Flash-GGUF:UD-Q4_K_XL ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf unsloth/GLM-5.3-Flash-GGUF:UD-Q4_K_XL ``` Other versions: ud-q2_k_xl (~96.5 GB) `llama cli -hf unsloth/GLM-5.3-Flash-GGUF:UD-Q2_K_XL` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run GLM-5.3-Flash (ud-q4_k_xl, needs ~192.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF/tree/main)) ```sh lms get https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF@UD-Q4_K_XL -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name Other versions: ud-q2_k_xl (~96.5 GB) `lms get https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF@UD-Q2_K_XL -y` Model repository and documentation: [huggingface.co/zai-org](https://huggingface.co/zai-org/GLM-5.3-Flash) [interactive version](https://whatmodel.sh/model/glm-5-3-flash) · [JSON](https://whatmodel.sh/api?format=json&all=1)