Compare / Mistral AI vs Z.ai
Mistral AI vs Z.ai: which AI lab is ahead?
Each lab's best model on sale goes head to head on every kind of work we measure, from our own benchmarks and licensed sources. 24 models in all; a lab is only as good as the model you would actually pick from it.
Kinds of work led by each lab's best model there, of the 9 both are measured on.
Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331). Models on sale means listed on OpenRouter today.
Mistral AI or Z.ai? The short answer
Z.ai is ahead: its best models beat Mistral AI's in 6 of the 9 kinds of work both are measured on; Mistral AI leads in 3.
- Pick Mistral AI for product listings, sticking to the facts and speed under pressure.
- Pick Z.ai for decks from an analysis, user surveys, marketing planning, agents and tool use, reasoning and knowledge and writing.
- On our overall leaderboard, GLM-5.3 is #30 (35 of 100) and Mistral Large 4 is #35 (24).
- Prices start at $0.058 per million tokens for Mistral AI (Mistral Small 3) and $0.15 for Z.ai (GLM-4.7-Flash).
Head to head
Each line names the model behind the number. Prices are blended, three input tokens to one output, at the developer's list price where we have read it; see every model's price. Intelligence Index from Artificial Analysis; overall score from our leaderboard.
Which lab is better for what
For each kind of work, each lab's best model there and its rank among every model on sale. Further right is better; the better rank is in green.
Every model, by intelligence and price
Each dot is a model on sale with an Intelligence Index from Artificial Analysis. Up is smarter; left is cheaper. Hover or tap a dot for its name.
Mistral AIZ.ai
On our overall leaderboard
The 5 of their models we rank overall: one score out of 100 across everything we publish (how the score works). Only models with enough of our own results are ranked.
- #30GLM-5.335
- #32GLM-5.3 FlashX Legal29
- #35Mistral Large 424
- #37Mistral Medium 3.515
- #38GLM-5V-Turbo12
Benchmark by benchmark
The headline result of every benchmark both labs have models on: each lab's best model and its rank among the models on sale. Green marks the better rank.
| Measure | Mistral AI | Z.ai |
|---|---|---|
| Agents and tool use | ||
| Berkeley Function Calling Leaderboard (BFCL) V4 · Overall accuracy · BFCL% correct, higher is better | 37.7%Mistral Medium 3 · 19 of 36 | 72.4%GLM-4.6 · 3 of 36 |
| Artificial Analysis · Terminal-Bench 4.0% of tasks, higher is better | 24.7%Mistral Large 4 · 25 of 107 | 41.9%GLM-5.3 · 13 of 107 |
| Artificial Analysis · Terminal-Bench 2.1% of tasks, higher is better | 50.6%Mistral Medium 3.5 · 59 of 107 | 84.3%GLM-5.3-Flash · 14 of 107 |
| Artificial Analysis · τ-bench banking% of tasks, higher is better | 15.1%Mistral Medium 3.5 · 56 of 104 | 50.3%GLM-5.3 · 3 of 104 |
| Decks from an analysis | ||
| DeckBench · Deck ratingrating, higher is better | 595Mistral Large 4 · 33 of 35 | 809GLM-5.3 FlashX Legal · 27 of 35 |
| Marketing planning | ||
| ROASBench · Overall scorescore out of 100, higher is better | 37.0Mistral Large 4 · 26 of 35 | 46.2GLM-5.3 FlashX Legal · 17 of 35 |
| Product listings | ||
| CatalogBench · Reliably publish-ready% of products, higher is better | 17.9%Mistral Medium 3.5 · 29 of 35 | 1.8%GLM-5V-Turbo · 32 of 35 |
| CatalogBench · Reliably publish-ready · sales brief% of products, higher is better | 10.7%Mistral Medium 3.5 · 20 of 35 | 0.0%GLM-5.3 FlashX Legal · 30 of 35 |
| Reasoning and knowledge | ||
| Artificial Analysis · Artificial Analysis Intelligence Indexindex score, higher is better | 38.4Mistral Large 4 · 34 of 185 | 44.8GLM-5.3 · 17 of 185 |
| Artificial Analysis · Humanity's Last Exam% of questions, higher is better | 35.0%Mistral Large 4 · 61 of 186 | 42.3%GLM-5.3 · 37 of 186 |
| Artificial Analysis · GPQA Diamond% of questions, higher is better | 76.9%Mistral Small 4 · 104 of 168 | 91.7%GLM-5.3 · 24 of 168 |
| Speed under pressure | ||
| BulletBench · Ladder Elo · Lightning 10+1ladder Elo, higher is better | 571Mistral Small 4 · 5 of 18 | 310GLM-5.3 FlashX Legal · 12 of 18 |
| BulletBench · Ladder Elo · Bullet 60sladder Elo, higher is better | 572Mistral Medium 3.5 · 9 of 36 | 358GLM-5.3 FlashX Legal · 20 of 36 |
| Sticking to the facts | ||
| Vectara Hallucination Leaderboard · Hallucination rate · Vectara% of summaries, lower is better | 5.1%Mistral Small 3 · 5 of 71 | 9.3%GLM-4.7-Flash · 34 of 71 |
| Arena (formerly LMArena) · Overall · Text factualityArena rating, higher is better | 1,434Mistral Large 3 · 72 of 109 | 1,467GLM-5.3 · 25 of 109 |
| User surveys | ||
| SurveyBench · Analysis ratingrating, higher is better | 770Mistral Large 4 · 19 of 26 | 937GLM-5.3 FlashX Legal · 16 of 26 |
| Writing | ||
| Arena (formerly LMArena) · Overall · TextArena rating, higher is better | 1,427Mistral Medium 3.5 · 74 of 153 | 1,478GLM-5.3 · 22 of 153 |
| UGI Leaderboard · Writing score · UGIscore out of 100, higher is better | 45.5Mistral Medium 3.5 · 64 of 124 | 67.0GLM-5.2 · 18 of 124 |
Their models, one against one
Quick answers
Is Mistral AI or Z.ai better?
Z.ai is ahead: its best models beat Mistral AI's in 6 of the 9 kinds of work both are measured on; Mistral AI leads in 3.
What is Mistral AI's best model?
Mistral Large 4: #35 on the Spring Prompt overall leaderboard and 38.4 on the Artificial Analysis Intelligence Index.
What is Z.ai's best model?
GLM-5.3: #30 on the Spring Prompt overall leaderboard and 44.8 on the Artificial Analysis Intelligence Index.
Which is better for agents and tool use, Mistral AI or Z.ai?
Z.ai: its best model there, GLM-5.3, ranks 15 of 159 models on sale, against 37 for Mistral AI's Mistral Large 4.
Which is better for reasoning and knowledge, Mistral AI or Z.ai?
Z.ai: its best model there, GLM-5.3, ranks 31 of 187 models on sale, against 54 for Mistral AI's Mistral Large 4.
Which is better for writing, Mistral AI or Z.ai?
Z.ai: its best model there, GLM-5.3, ranks 24 of 169 models on sale, against 85 for Mistral AI's Mistral Medium 3.5.
Which is cheaper, Mistral AI or Z.ai?
Mistral AI's cheapest model with results, Mistral Small 3, costs $0.058 per million tokens (three input to one output), against $0.15 for GLM-4.7-Flash from Z.ai.
Other labs compared
- Mistral AI vs OpenAIBest model against best model
- Anthropic vs Mistral AIBest model against best model
- Google vs Mistral AIBest model against best model
- Mistral AI vs xAIBest model against best model
- Meta vs Mistral AIBest model against best model
- DeepSeek vs Mistral AIBest model against best model
- Alibaba vs Mistral AIBest model against best model
- OpenAI vs Z.aiBest model against best model
Which model is best for your product?
Benchmarks aren't your task. Run models from Mistral AI, Z.ai and the rest on your own prompt, and see the cheapest one that passes.