GLM-5.1
Z.AIOpen weightsReasoningVerifiedReleased Apr 7, 2026Input: text
Overall score
56.3
Rank 63 of 222 · 45.6–67.0
Context window
203K
203,000 tokens
Price per 1M tokens
$1.40 in · $4.40 out
Source: BenchLM
Blended price
$2.15
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Instruction following#693.8
- 2Reasoning75.6
- 3Math#766.7
- 4Agentic & tool use56.3
- 5Coding55.7
- 6Knowledge51.9
Benchmark results (20)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| Gert Labs | 60.11 | #12 of 52 |
| BrowseComp | 68.0 | #33 of 50 |
| MCP Atlas | 71.8 | #23 of 42 |
| Terminal-Bench 2.0 | 63.5 | #14 of 39 |
| Claw-Eval | 62.3 | #19 of 36 |
| HLE w/ tools | 52.3 | #19 of 27 |
| CyberGym | 68.7 | #19 of 25 |
| τ³-bench results | 70.6 | #6 of 23 |
| ResearchClawBench | 18.2 | #7 of 19 |
Math
| FrontierMath v2 (Tiers 1-3) | 36.842 | #47 of 64 |
| AIME26 | 95.3 | #11 of 29 |
| HMMT Feb 2026 | 82.6 | #18 of 24 |
| HMMT Nov 2025 | 94.0 | #4 of 12 |
| MMAnswerBench | 83.8 | #4 of 10 |
Coding
| SWE-bench Pro | 58.4 | #33 of 69 |
| NL2Repo | 42.7 | #20 of 31 |
| SWE-Rebench | 62.7 | #3 of 13 |
| OpenHarmony Bench | 52.3 | #10 of 12 |
Multimodal
| Design Arena Website | 1290 | #19 of 91 |
Knowledge
| HLE | 31.0 | #40 of 69 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price (BenchLM) | $1.40 | $4.40 | — |
| OpenRouter z-ai/glm-5.1 | $0.97 | $3.04 | $0.18 |
Z.AI's official pricing page lists GLM-5.1 at $1.40 input / $4.40 output per million tokens.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text | 1462 | #39 · 62.9K votes |
| WebDev (coding) | 1508 | #60 · 10.7K votes |
Open weights
From the Hugging Face repository.
- Parameters
- 754B
- Licence
- mit
- Downloads (30 days)
- 554.5K
- Repository
- zai-org/GLM-5.1 ↗
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
- GLM-554.9
- GLM-5-Turbo53.2
Similar score
- Kimi K2.5 (Reasoning)56.2 · $1.20
- Claude Sonnet 4.655.9 · $6.00
- Qwen3.5 397B55.4 · $1.35
- GPT-5.157.2 · $3.44
- GPT-5.2-Codex55.4 · $4.81