GLM-5
- In Input modalities: Text Out Output modalities: Text
- Reasoning
- Reasoning can be off
- Tool use
- Structured output
- Open weights
Prices
2 offerings · prices as listed| Conditions/ 1M tokens | Input | Cache read | Output |
|---|---|---|---|
| Standard | $1.00 | $0.20 | $3.20 |
Prices from Pricing (opens in a new tab) (checked Nov 8, 2026)
| Conditions/ 1M tokens | Input | Cache read | Output |
|---|---|---|---|
| Standard | ¥4.00 | ¥1.00 | ¥18.00 |
| Prompts over 32K tokens | ¥6.00 | ¥1.50 | ¥22.00 |
Prices from 智谱 BigModel：API 定价 (opens in a new tab) (checked Nov 9, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Licence
- Apache-2.0
- Family
- glm
- Model ID
Not stated on Zhipu AI’s pages: max input, reasoning efforts, max thinking budget, built-in tools, knowledge cutoff, aliases.
Run locally
Weights: zai-org/GLM-5- Parameters 753.9B
- Active 51.7B (estimated) · 256 experts, 8 per token
- Native precision BF16
- Layers 78
- KV cache per token 88 KB
- Config positions 198K (config.json max_position_embeddings)
- API context 195K (maker’s API)
- Needs
- ≈ 392.3 GB (weights 373.0 + KV cache 0.7 + overhead 18.6)
- Usable
- 21.6 GB (90% of its memory)
- Try
- it’s too large for this hardware; see the smallest that runs it below
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| BF16 (native) | 1404.2 GB | ≈ 1475.1 | ✕Doesn’t fit | — |
| Q8_0 | 746.0 GB | ≈ 784.0 | ✕Doesn’t fit | — |
| FP8maker’s file | 704.2 GB | ≈ 740.1 | ✕Doesn’t fit | — |
| Q6_K | 576.6 GB | ≈ 606.1 | ✕Doesn’t fit | — |
| Q5_K_M | 499.4 GB | ≈ 525.0 | ✕Doesn’t fit | — |
| Q5_0 | 482.7 GB | ≈ 507.5 | ✕Doesn’t fit | — |
| Q4_K_M | 426.5 GB | ≈ 448.5 | ✕Doesn’t fit | — |
| Q4_0 | 394.9 GB | ≈ 415.4 | ✕Doesn’t fit | — |
| IQ4_XS | 373.0 GB | ≈ 392.3 | ✕Doesn’t fit | — |
| Q3_K_Mquality drops | 343.1 GB | ≈ 361.0 | ✕Doesn’t fit | — |
| IQ3_XXSquality drops | 268.5 GB | ≈ 282.7 | ✕Doesn’t fit | — |
| Q2_Kquality drops | 263.3 GB | ≈ 277.1 | ✕Doesn’t fit | — |
| IQ2_XXSquality drops | 180.8 GB | ≈ 190.5 | ✕Doesn’t fit | — |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): no single listed machine; about 7 × 80 GB GPUs.
Run commands
The maker publishes no files at this precision. Choose its native precision or one of its own builds, or quantise it yourself with llama.cpp.
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-11. How it’s estimated
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 146.4
#73 of 118 listed models · CI 144.9–148
- LMArena Text 1,458
#43 of 163 listed models · CI 1,453–1,462 · 29,018 votes
- Epoch AI · GPQA Diamond 87.8%
#56 of 168 listed models · ±2.34
- Epoch AI · OTIS Mock AIME 2024–2025 80%
#82 of 159 listed models · ±6
- Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 7 – Nov 5, 2026
- LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Nov 8, 2026
- EQ-Bench by Samuel J. Paech (eqbench.com; EQ-bench/EQ-bench-site on GitHub), MIT License. · License · as of Oct 28, 2026
- changes we made
Provider-reported scores
Published by Zhipu AI, under the conditions it states| Benchmark | Setting | Score |
|---|---|---|
| Humanity's Last Exam | text-only subset, no tools, 131,072 max generation tokens | 30.5% |
| text-only subset, with tools, 202,752 max context | 50.4% | |
| GPQA Diamond | temperature 1.0, 131,072 max generation tokens | 86% |
| SWE-bench Verified | OpenHands harness, 200K context | 77.8% |
| Terminal-Bench 2.0 | Terminus 2 harness, 2h timeout, 128K context | 56.2% |
| Claude Code 2.1.14 harness, think mode, avg of 5 runs | 56.2% | |
| BrowseComp | without context management (last 5 turns kept) | 62% |
| with discard-all context management | 75.9% | |
| τ²-bench | with vendor prompt adjustments (Retail, Telecom) and Airline domain fixes | 89.7% |
| MCP-Atlas | 500-task public subset, think mode, Gemini 3 Pro judge | 67.8% |
Included in coding plans
- Model Studio Coding Plan Pro $50.00 / month6,000 requests every 5 hours45,000 requests per week90,000 requests per month
- Kiro Free Free50 credits per month
- Kiro Pro $20.00 / month1,000 credits per month
- Kiro Pro+ $40.00 / month2,000 credits per month
- Kiro Pro Max $100.00 / month5,000 credits per month
- Kiro Power $200.00 / month10,000 credits per month
Changes
All changes →No changes recorded yet.
Sources
- zai-org/GLM-5 model card (opens in a new tab) huggingface.co · checked Nov 6, 2026
- Pricing (opens in a new tab) docs.z.ai · checked Nov 8, 2026
- 智谱 BigModel：API 定价in Chinese (opens in a new tab) docs.bigmodel.cn · checked Nov 9, 2026
- GLM-5 (opens in a new tab) docs.z.ai · checked Nov 8, 2026
- New Released (opens in a new tab) docs.z.ai · checked Nov 8, 2026
- zai-org/GLM-5 model card (opens in a new tab) huggingface.co · checked Nov 8, 2026
- models.dev: glm-5 (community data, MIT) (opens in a new tab) github.com · checked Sep 30, 2026
- zai-org/GLM-5 model card (opens in a new tab) huggingface.co · checked Nov 8, 2026
- GitHub - zai-org/GLM-5: GLM-5: From Vibe Coding to Agentic Engineering · GitHub (opens in a new tab) github.com · checked Nov 5, 2026
- zai-org/GLM-5 config.json (opens in a new tab) huggingface.co · checked Nov 4, 2026
- zai-org/GLM-5-FP8 files (opens in a new tab) huggingface.co · checked Nov 5, 2026
Zhipu AI’s official policies: data use, privacy, terms →
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
Zhipu AI · China · z.ai
Related models
Snapshots, the same family and other models from this provider