Grok Build 0.1 vs GLM 5
Grok Build 0.1 (xAI, released 2026-05-20) and GLM 5 (Z.ai, released 2026-02-11) both target agentic software engineering, but they differ in modality, price, and how much benchmark data backs them up. Grok Build 0.1 accepts text and images with a 256K context window; GLM 5 is text-only with 202,752 tokens of context. Grok Build 0.1 forces reasoning mode on every call, GLM 5 makes it optional but on by default. This comparison uses actual pricing and Design Arena scores, not guesses.
Spec vs spec
| Spec | Grok Build 0.1 | GLM 5 |
|---|---|---|
| Context window | 256K | 203K |
| Input modalities | text, image | text |
| Output modalities | text | text |
| Released | May 20, 2026 | Feb 11, 2026 |
| Reasoning | always on | optional |
Pricing
Per 1M tokens. Provider price plus the flat 3% fee — the sum is what you pay.
x-ai/grok-build-0.1
Input · 1M tokens
$1.00 + 3%$1.03
Output · 1M tokens
$2.00 + 3%$2.06
Cache read · 1M tokens
$0.200 + 3%$0.206
FEE — FLAT, EVERY MODEL3%
z-ai/glm-5
Input · 1M tokens
$0.600 + 3%$0.618
Output · 1M tokens
$1.92 + 3%$1.98
Cache read · 1M tokens
$0.120 + 3%$0.124
FEE — FLAT, EVERY MODEL3%
One workload, priced on both
10M input + 2M output tokens at each model's price, flat 3% fee included.
x-ai/grok-build-0.1
$14.42
$14.00 provider + 3%
z-ai/glm-5Cheaper
$10.14
$9.84 provider + 3%
Pricing math
Grok Build 0.1 charges $1.00/M input and $2.00/M output from xAI; on OpenKey that's $1.03/M input and $2.06/M output after the flat 3% fee. GLM 5 charges $0.60/M input and $1.92/M output from Z.ai, or $0.618/M and $1.9776/M on OpenKey.
On a workload of 10M input tokens + 2M output tokens: Grok Build 0.1 costs $14.00, GLM 5 costs $9.84. That's a 1.67x gap on input price alone, and it holds up across the full workload — GLM 5 is meaningfully cheaper for the same token volume. If you're running high-volume agent loops with lots of tool calls and retries, that difference compounds fast.
Coding and agent benchmarks
GLM 5 has Design Arena scores across 12 categories. In the agents arena it ranks #3 in Godot game dev (elo 1237, 54.8% win rate), #6 in Android native (elo 1244, 62% win rate), #10 in mobile apps (elo 1222, 53.1% win rate), and #13 in fullstack (elo 1190, 52.8% win rate). Across the models arena it lands mid-pack: #15 in gamedev and 3D, #16 in code categories, #18 in website, #19 in SVG, #21 in UI components, #22 in dataviz, #26 in ASCII art.
Grok Build 0.1 has no published Design Arena or artificial_analysis benchmarks in this dataset, so there's no independent score to compare against GLM 5's numbers. That's not a knock on the model — it's just untested here.
Context and modality
Grok Build 0.1 has the larger context window at 256,000 tokens versus GLM 5's 202,752 — a 1.26x ratio. That extra room matters for large codebases or long agent traces, but the gap is modest, not a tier difference.
The bigger split is input modality. Grok Build 0.1 accepts text and images; GLM 5 is text-only. If your workflow includes screenshots, diagrams, or UI mockups as input, GLM 5 is off the table regardless of price. Neither model publishes a knowledge cutoff date, so don't assume either has current-events awareness — test it directly if that matters for your use case.
When to pick each
Pick GLM 5 when cost per token matters and you're doing pure-text agentic coding — its cheaper rate and its Design Arena rankings in game dev and Android agent tasks make it a reasonable default for backend and fullstack work. Pick Grok Build 0.1 when you need to feed the model screenshots or images alongside code, or when the extra 53,248 tokens of context (256K vs 202,752) matters for your document size. Both models are available on OpenKey with one key and the same flat 3% fee on top of provider pricing, so switching between them for A/B testing doesn't cost you anything beyond the token price difference.
Which model for which job
| Use case | Pick | Why |
|---|---|---|
| High-volume agent loops (cost-sensitive) | GLM 5 | $9.84 vs $14.00 on a 10M-in/2M-out workload |
| Coding tasks with image/screenshot input | Grok Build 0.1 | GLM 5 is text-only; Grok Build 0.1 supports text+image input |
| Android native or Godot game dev agents | GLM 5 | Ranks #6 (Android, elo 1244) and #3 (Godot, elo 1237) on Design Arena |
| Very large codebase context | Grok Build 0.1 | 256,000 token context vs GLM 5's 202,752 |
| Need independently benchmarked coding scores | GLM 5 | Has 12 published Design Arena category scores; Grok Build 0.1 has none in this dataset |
Questions
- Which model is cheaper to run at scale?
- GLM 5. On a 10M input / 2M output workload it costs $9.84 total versus $14.00 for Grok Build 0.1 — driven mostly by GLM 5's $0.60/M input price against Grok's $1.00/M, a 1.67x gap.
- Does either model support image input?
- Only Grok Build 0.1. It accepts text and image inputs with text output. GLM 5 is text-to-text only, so any workflow involving screenshots or visual context needs Grok Build 0.1.
- Which has the bigger context window?
- Grok Build 0.1, at 256,000 tokens versus GLM 5's 202,752 — a 1.26x ratio. The difference is real but not dramatic; either handles long documents reasonably well.
- Is GLM 5 actually better at coding, or just cheaper?
- GLM 5 has published Design Arena data showing rank #3 in Godot game dev (elo 1237) and #6 in Android native (elo 1244). Grok Build 0.1 has no benchmark scores in this dataset, so there's nothing to directly compare its coding quality against.