Mistral Large 3 vs GLM-4.5
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by Mistral
|
by Z.AI
|
|---|---|---|
| Overview | ||
| Status | Current mid tier | Current mid tier Open weights |
| Released | Nov 1, 2025 | Jan 1, 2025 |
| Pricing per million tokens | ||
| Input | $0.500/Mtok | $0.600/Mtok |
| Output | $1.50/Mtok | $2.20/Mtok |
| Blended avg | $1.00/Mtok | $1.40/Mtok |
| Cached input | — | $0.110/Mtok |
| Specifications | ||
| Context window | 128K tokens | 128K tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
text
|
text
|
| Benchmarks sources: Mistral model card, Artificial Analysis / LMSYS Chatbot Arena (UC Berkeley) | ||
| Independent composite scores measured for both models — each on its own scale | ||
| LMSYS Chatbot Arena | 1415.3 ELO | 1410.7 ELO |
| Per-benchmark results published for Mistral Large 3 only — no independent per-benchmark scores exist for GLM-4.5 yet | ||
| Avg benchmark score | 53.6 | — |
| Perf / dollar | 53.6 | — |
| GPQA Diamond | 58 | — |
| SWE-Bench Verified | 35 | — |
| Humanity's Last Exam | 18 | — |
| ARC-AGI 2 | 20 | — |
| AIME 2025 | 52 | — |
| MMMLU | 78 | — |
| BFCL | 68 | — |
| HumanEval | 85 | — |
| MATH 500 | 68 | — |
| Providers | ||
| Available from |
Mistral — $0.500/$1.50/Mtok
|
Z.AI — $0.600/$2.20/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | Mistral Large 3 | GLM-4.5 | Savings |
|---|---|---|---|
| 1M tokens | $1 | $1.4 | $0.4 (28.6%) |
| 10M tokens | $10 | $14 | $4 (28.6%) |
| 100M tokens | $100 | $140 | $40 (28.6%) |
| 1000M tokens | $1000 | $1400 | $400 (28.6%) |
The cheaper option here — Mistral Large 3 costs $1.00/Mtok blended on Mistral.
Summary
Mistral Large 3 by Mistral costs $0.500/Mtok input and $1.50/Mtok output, with a 128K-token context window. It supports text input.
GLM-4.5 by Z.AI costs $0.600/Mtok input and $2.20/Mtok output, with a 128K-token context window. It supports text input.
On a blended cost basis, Mistral Large 3 is 28.6% cheaper than GLM-4.5.
The two aren't directly comparable on average benchmark score: Mistral Large 3 has published per-benchmark results, while GLM-4.5 does not yet — it is measured today on independent composites (see the table above).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.