GLM-5.1 NEW
Z.ai's largest open MoE reasoning model
Z.ai's GLM-5.1 in reasoning mode — 744B total / 40B active per token MoE under MIT license, 200K context, ~Top 10% on AA's Intelligence Index. Competes with DeepSeek V4 Pro and Kimi K2 Thinking in the open-weights frontier tier.
Intelligence
Good
Speed
Slow
60 tok/s output
Cost
Moderate
$1.40 in / $4.40 out
Context
200K
Up to 200,000 tokens
How are Intelligence, Speed & Cost bucketed?
Intelligence and Speed buckets are percentile ranks on
Artificial Analysis. Cost buckets are fixed dollar
thresholds keyed off output-token price ($/M out).
Intelligence
- Top 1%≤ 1%
- Top 5%≤ 5%
- Top 10%≤ 10%
- Good≤ 25%
- Medium≤ 50%
- Below avg> 50%
Speed
- Top 1%≥ 345 tok/s
- Top 5%≥ 237 tok/s
- Top 10%≥ 196 tok/s
- Good≥ 146 tok/s
- Medium≥ 90 tok/s
- Slow< 90 tok/s
Cost
- Freeopen weights · self-host
- Low< $1 / M out
- Moderate$1–5 / M out
- High≥ $5 / M out
Why it matters
Marks the moment open-weights frontier crosses Top 10% on AA's Intelligence Index — the first time a permissive-license model has reached this tier of benchmark performance.
Core Capabilities
Long Documents
Handles entire codebases, books, and multi-doc RAG.
Generative
Produces images, video, audio, or other media.
Reasoning
Solves complex math, logic, and planning tasks.
Context Window
200k tokens
≈ 154+ pages
4k Chat 聊天
32k Long docs 长文档
128k Books 整本书
400k Multi-doc 多文档
1M Codebase 整个代码库
10M
200k
Availability
API
Available
Product / App
Not available
Open Source
Released
Enterprise
Contact sales
Pricing Model
Free / self-host
Open weights — pay only for compute.
Self-host Capability / Performance
Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).
Lower 20% Upper 80% This model
Context / memory
Context window size · log-scaled
6.0
9.0
6.7
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this
Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.
What it feels like
- Reasoning-style traces, frontier-tier benchmarks (~Top 10% on AA Intelligence Index)
- 744B/40B-active MoE — open-weights peer to DeepSeek V4 Pro and Kimi K2 Thinking
Best use cases
- Hardest reasoning tasks among open-weights models — GPQA 86.8%, HLE 28%
- Self-host scenarios where you can run a 744B/40B-active MoE
- API price-sensitive serving at $1.40/$4.40 — cheaper than Claude / GPT frontier tiers
Tools to try
Not ideal for
- Multimodal / vision tasks — text-only release
- Single-GPU serving — needs sharded inference
- Latency-sensitive UX — reasoning chains add output time
Model Evolution
GLM is Zhipu's language model family.