Models, measured side by side
GLM 5.3 vs Grok 4.7
Compare GLM 5.3, Grok 4.7 by token pricing, context length, benchmark results, speed and tool support.
Monthly workload
Monthly cost breakdown
InputOutput
GLM 5.3
$45 / mo
Grok 4.7
$70 / mo
Performance comparison
GLM 5.3
44.8
Grok 4.7
46.4
Overview
Pricing & capacity
| Pricing & capacity | GLM 5.3 | Grok 4.7 |
|---|---|---|
| Input / 1M tokens | $1.40 | $2.00 |
| Output / 1M tokens | $3.39 | $6.00 |
| Cached input / 1M | $0.05 | $0.50 |
| Monthly cost | $45 | $70 |
| Context window | 1M | 500K |
| Maximum output | 944K | 450K |
Input / 1M tokens
- GLM 5.3
- $1.40
- Grok 4.7
- $2.00
Output / 1M tokens
- GLM 5.3
- $3.39
- Grok 4.7
- $6.00
Cached input / 1M
- GLM 5.3
- $0.05
- Grok 4.7
- $0.50
Monthly cost
- GLM 5.3
- $45
- Grok 4.7
- $70
Context window
- GLM 5.3
- 1M
- Grok 4.7
- 500K
Maximum output
- GLM 5.3
- 944K
- Grok 4.7
- 450K
Benchmarks & performance
| Benchmarks & performance | GLM 5.3 | Grok 4.7 |
|---|---|---|
| Arena Elo | Not listed | Not listed |
| MMLU-Pro | Not listed | Not listed |
| Intelligence Index | 44.8 | 46.4 |
| GPQA | 91.7% | Not listed |
| Humanity’s Last Exam | 42.3% | 43.1% |
| SciCode | 59% | 57.4% |
| Output speed | 72 tokens/s | 83 tokens/s |
| Time to first token | 2.73 s | 43.17 s |
| DesignArena rating | Not listed | 1,269 |
| DesignArena win rate | Not listed | 47.6% |
Arena Elo
- GLM 5.3
- Not listed
- Grok 4.7
- Not listed
MMLU-Pro
- GLM 5.3
- Not listed
- Grok 4.7
- Not listed
Intelligence Index
- GLM 5.3
- 44.8
- Grok 4.7
- 46.4
GPQA
- GLM 5.3
- 91.7%
- Grok 4.7
- Not listed
Humanity’s Last Exam
- GLM 5.3
- 42.3%
- Grok 4.7
- 43.1%
SciCode
- GLM 5.3
- 59%
- Grok 4.7
- 57.4%
Output speed
- GLM 5.3
- 72 tokens/s
- Grok 4.7
- 83 tokens/s
Time to first token
- GLM 5.3
- 2.73 s
- Grok 4.7
- 43.17 s
DesignArena rating
- GLM 5.3
- Not listed
- Grok 4.7
- 1,269
DesignArena win rate
- GLM 5.3
- Not listed
- Grok 4.7
- 47.6%
Tools & features
| Tools & features | GLM 5.3 | Grok 4.7 |
|---|---|---|
| Function calling | Supported | Supported |
| Structured outputs | Supported | Supported |
| JSON mode | Supported | Supported |
| Reasoning | Supported | Supported |
| Built-in web search | Not listed | Not listed |
| Log probabilities | Supported | Supported |
| Deterministic seed | Supported | Supported |
| Parallel tool calls | Supported | Not listed |
| Prompt caching | Supported | Supported |
Function calling
- GLM 5.3
- Supported
- Grok 4.7
- Supported
Structured outputs
- GLM 5.3
- Supported
- Grok 4.7
- Supported
JSON mode
- GLM 5.3
- Supported
- Grok 4.7
- Supported
Reasoning
- GLM 5.3
- Supported
- Grok 4.7
- Supported
Built-in web search
- GLM 5.3
- Not listed
- Grok 4.7
- Not listed
Log probabilities
- GLM 5.3
- Supported
- Grok 4.7
- Supported
Deterministic seed
- GLM 5.3
- Supported
- Grok 4.7
- Supported
Parallel tool calls
- GLM 5.3
- Supported
- Grok 4.7
- Not listed
Prompt caching
- GLM 5.3
- Supported
- Grok 4.7
- Supported
API & availability
| API & availability | GLM 5.3 | Grok 4.7 |
|---|---|---|
| API identifier | glm-5.3 | grok-4.7 |
| API providers | Not listed | xAI, xAI, xAI, xAI, xAI |
| Knowledge cutoff | Not listed | Not listed |
| Prices checked | Oct 8, 2026 | Oct 8, 2026 |
API identifier
- GLM 5.3
glm-5.3- Grok 4.7
grok-4.7
API providers
- GLM 5.3
- Not listed
- Grok 4.7
- xAI, xAI, xAI, xAI, xAI
Knowledge cutoff
- GLM 5.3
- Not listed
- Grok 4.7
- Not listed
Prices checked
- GLM 5.3
- Oct 8, 2026
- Grok 4.7
- Oct 8, 2026
About the models
GLM 5.3
GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...
Full pricing & detailsGrok 4.7
Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...
Full pricing & detailsComparison FAQ
GLM 5.3: $1.40 input and $3.39 output per million tokens. Grok 4.7: $2.00 input and $6.00 output per million tokens. The cheaper choice depends on your input-to-output ratio; a model can have a lower input rate but a higher output rate.