GLM 5.3 Flash · Data analysis

GLM 5.3 Flash for data analysis: our test runs

GLM 5.3 Flash ran our 5 data analysis prompts in llmwise, through the same pipeline your messages take. Here's every reply as it came, whether it passed, what it cost and what it counts as on Pro, next to the other models.

Start free See pricing

Based on 5 of our test runs on September 27, 2026, through OpenRouter with the app's own prompt and settings. Updated October 15, 2026.

Short answer

In our test runs on September 27, 2026, GLM 5.3 Flash passed 4 of 5 of our data analysis prompts (1 of the 2 hard ones): tied 17th of 20 models. Each reply counted as 1 message on Pro, where it has 60 a day, and cost $0.0004 to run on average.

Prompt by prompt

Each of the 5 data analysis prompts, and how GLM 5.3 Flash's reply did.

GLM 5.3 Flash on each data analysis prompt
Prompt	Result	Cost	Time
The region with the most revenueEveryday	Passed: Final answer North, with 1,085.00 in revenue: right.	$0.0005	8.8 s
Average order value in AugustEveryday	Passed: Final answer 300.00: right.	$0.0004	14.5 s
Revenue change from July to AugustEveryday	Passed: Final answer 67.4%: right.	$0.0004	9.8 s
A median, filtered two waysHard	Passed: Final answer 6.25: right.	$0.0001	3.7 s
Correlation between ad spend and sign-upsHard	Failed: Final answer 0.99; expected 0.97.	$0.0005	19.4 s

Every data analysis reply from GLM 5.3 Flash

The region with the most revenue

An everyday prompt: “The region with the most revenue”, in full.

Passed: Final answer North, with 1,085.00 in revenue: right.

Revenue per line (quantity × price):
- 1001 West: 2×45 = 90
- 1002 East: 1×120 = 120
- 1003 West: 1×310 = 310
- 1004 North: 4×45 = 180
- 1005 East: 3×45 = 135
- 1006 North: 2×120 = 240
- 1007 West: 2×120 = 240
- 1008 East: 1×310 = 310
- 1009 North: 2×310 = 620
- 1010 West: 5×45 = 225
- 1011 East: 3×120 = 360
- 1012 North: 1×45 = 45

Totals by region:
- West: 90 + 310 + 240 + 225 = 865
- East: 120 + 135 + 310 + 360 = 925
- North: 180 + 240 + 620 + 45 = 1,085

Final answer: North, with $1,085.00 in revenue

655 tokens in, 478 out (220 of them reasoning) · 8.8 s · $0.0005 · 1 message on Pro · answered by z-ai/glm-5.3-flash via Wafer · September 27, 2026

Average order value in August

An everyday prompt: “Average order value in August”, in full.

Passed: Final answer 300.00: right.

**August 2026 orders (1007–1012):**

| Order | Qty × Price | Revenue |
|-------|-------------|---------|
| 1007 | 2 × 120.00 | 240.00 |
| 1008 | 1 × 310.00 | 310.00 |
| 1009 | 2 × 310.00 | 620.00 |
| 1010 | 5 × 45.00 | 225.00 |
| 1011 | 3 × 120.00 | 360.00 |
| 1012 | 1 × 45.00 | 45.00 |

Total revenue = 240 + 310 + 620 + 225 + 360 + 45 = **1,800.00**
Number of orders = **6**
Average = 1,800.00 ÷ 6 = **300.00**

Final answer: 300.00

662 tokens in, 372 out (171 of them reasoning) · 14.5 s · $0.0004 · 1 message on Pro · answered by z-ai/glm-5.3-flash via Wafer · September 27, 2026

Revenue change from July to August

An everyday prompt: “Revenue change from July to August”, in full.

Passed: Final answer 67.4%: right.

**July 2026 revenue:**
- 2×45 = 90
- 1×120 = 120
- 1×310 = 310
- 4×45 = 180
- 3×45 = 135
- 2×120 = 240
- Total = **1075**

**August 2026 revenue:**
- 2×120 = 240
- 1×310 = 310
- 2×310 = 620
- 5×45 = 225
- 3×120 = 360
- 1×45 = 45
- Total = **1800**

**Percentage change:** (1800 − 1075) / 1075 × 100 = 725 / 1075 × 100 ≈ 67.4419% → 67.4%

Final answer: 67.4%

667 tokens in, 424 out (231 of them reasoning) · 9.8 s · $0.0004 · 1 message on Pro · answered by z-ai/glm-5.3-flash via Wafer · September 27, 2026

A median, filtered two ways
A hard prompt: “A median, filtered two ways”, in full.
Passed: Final answer 6.25: right.
```
Billing + high tickets: T1=5.5, T4=2, T6=9, T9=7, T10=26, T12=4

Sorted: 2, 4, 5.5, 7, 9, 26 — six values, so median = (5.5 + 7)/2 = 6.25

Final answer: 6.25
```
526 tokens in, 142 out (53 of them reasoning) · 3.7 s · $0.0001 · 1 message on Pro · answered by z-ai/glm-5.3-flash via AtlasCloud · September 27, 2026
Correlation between ad spend and sign-ups
A hard prompt: “Correlation between ad spend and sign-ups”, in full.
Failed: Final answer 0.99; expected 0.97.
```
Working:
- Mean ad_spend = 6150/8 = 768.75; mean signups = 434/8 = 54.25
- Σ(dx·dy) = 24428.125
- Σdx² = 624687.5; Σdy² = 971.5
- r = 24428.125 / √(624687.5 × 971.5) = 24428.125 / √606,871,906.25 ≈ 24428.125 / 24634.8 ≈ 0.9916

Final answer: 0.99
```
468 tokens in, 1,022 out (338 of them reasoning) · 19.4 s · $0.0005 · 1 message on Pro · answered by z-ai/glm-5.3-flash via Relace · September 27, 2026

How it compares on data analysis

GLM 5.3 Flash beside the job's picks among every model. 16 models share the hard-problems pick, and GLM 5.3 Flash isn't one of them.

GLM 5.3 Flash beside other models on data analysis
Model	Passed	On Pro	Cost per reply
GLM 5.3Our pick: hard problems (shared), best value	5 of 5	250 a month on Pro	$0.0019
GPT-6 LunaOur pick: hard problems (shared), everyday	5 of 5	60 a day on Pro	$0.0002
GLM 5.3 Flash	4 of 5	60 a day on Pro	$0.0004

On Pro: Pro's count on each model. Cost: what OpenRouter charged us per reply, on average.

Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Plus $3.75, Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.

GLM 5.3 Flash next to each model on data analysis

GLM 5.3 Flash against each model on the same 5 data analysis prompts. One run each, through OpenRouter: a wait depends on the provider and the load that day, so a lead under 10% counts as close.

Against Claude Fable 5.1: GLM 5.3 Flash answered 2.0× later (9.8 s to 4.9 s), cost 73.6× less, and passed 4 of 5 to its 5. Only Claude Fable 5.1 passed Correlation between ad spend and sign-ups.
Against Claude Opus 5.5: GLM 5.3 Flash answered 2.1× later (9.8 s to 4.6 s), cost 34.0× less, and passed 4 of 5 to its 5. Only Claude Opus 5.5 passed Correlation between ad spend and sign-ups.
Against Claude Sonnet 5.5: GLM 5.3 Flash answered 4.3× later (9.8 s to 2.3 s), cost 12.2× less, and passed 4 of 5 to its 5. Only Claude Sonnet 5.5 passed Correlation between ad spend and sign-ups.
Against Claude Sonnet 5: GLM 5.3 Flash answered 1.8× later (9.8 s to 5.3 s), cost 17.9× less, and passed 4 of 5 to its 4.
Against Claude Haiku 5.5: GLM 5.3 Flash answered 3.0× later (9.8 s to 3.3 s), cost within 10% of it, and passed 4 of 5 to its 5. Only Claude Haiku 5.5 passed Correlation between ad spend and sign-ups.
Against Claude Haiku 4.5: GLM 5.3 Flash answered 3.6× later (9.8 s to 2.7 s), cost 6.5× less, and passed 4 of 5 to its 3. Only GLM 5.3 Flash passed Average order value in August.
Against GPT-6 Astra: GLM 5.3 Flash answered 3.0× later (9.8 s to 3.3 s), cost 40.2× less, and passed 4 of 5 to its 5. Only GPT-6 Astra passed Correlation between ad spend and sign-ups.
Against GPT-6.1 Sol: GLM 5.3 Flash answered 3.3× later (9.8 s to 3.0 s), cost 4.1× less, and passed 4 of 5 to its 5. Only GPT-6.1 Sol passed Correlation between ad spend and sign-ups.
Against GPT-6 Sol: GLM 5.3 Flash answered 3.5× later (9.8 s to 2.8 s), cost 8.8× less, and passed 4 of 5 to its 5. Only GPT-6 Sol passed Correlation between ad spend and sign-ups.
Against GPT-6 Luna: GLM 5.3 Flash answered 3.9× later (9.8 s to 2.5 s), cost 2.2× more, and passed 4 of 5 to its 5. Only GPT-6 Luna passed Correlation between ad spend and sign-ups.
Against Gemini 4 Argon: GLM 5.3 Flash answered 1.4× later (9.8 s to 6.9 s), cost 7.4× less, and passed 4 of 5 to its 5. Only Gemini 4 Argon passed Correlation between ad spend and sign-ups.
Against Gemini 3.1 Pro (preview): GLM 5.3 Flash answered 1.1× later (9.8 s to 8.6 s), cost 40.2× less, and passed 4 of 5 to its 5. Only Gemini 3.1 Pro (preview) passed Correlation between ad spend and sign-ups.
Against Gemini 3.8 Flash: GLM 5.3 Flash answered 1.8× later (9.8 s to 5.4 s), cost 7.2× less, and passed 4 of 5 to its 4.
Against DeepSeek V4.1 Flash: GLM 5.3 Flash answered 11.3× later (9.8 s to 0.9 s), cost 2.2× less, and passed 4 of 5 to its 5. Only DeepSeek V4.1 Flash passed Correlation between ad spend and sign-ups.
Against DeepSeek V4 Pro: GLM 5.3 Flash answered 2.4× later (9.8 s to 4.1 s), cost 7.4× less, and passed 4 of 5 to its 5. Only DeepSeek V4 Pro passed Correlation between ad spend and sign-ups.
Against Grok 4.7: GLM 5.3 Flash answered within 10% of its time (9.8 s to 8.9 s), cost 28.9× less, and passed 4 of 5 to its 5. Only Grok 4.7 passed Correlation between ad spend and sign-ups.
Against Kimi K3: GLM 5.3 Flash answered 2.8× later (9.8 s to 3.5 s), cost 17.6× less, and passed 4 of 5 to its 5. Only Kimi K3 passed Correlation between ad spend and sign-ups.
Against GLM 5.3: GLM 5.3 Flash answered 5.1× later (9.8 s to 1.9 s), cost 5.0× less, and passed 4 of 5 to its 5. Only GLM 5.3 passed Correlation between ad spend and sign-ups.
Against Mistral Large 4: GLM 5.3 Flash answered 1.6× sooner (9.8 s to 15.8 s), cost 9.7× less, and passed 4 of 5 to its 5. Only Mistral Large 4 passed Correlation between ad spend and sign-ups.

How these runs were done

Final answer. Automatic. The reply's last “Final answer:” line must hold the right value.

How the runs were done, and every data analysis prompt.

GLM 5.3 Flash, and data analysis, elsewhere

Questions

Is GLM 5.3 Flash good for data analysis?

In our test runs it passed 4 of 5 data analysis prompts, tied 17th of the 20 models in llmwise. Every reply is on this page, so you can judge them yourself.

How many of my messages does a data analysis reply from GLM 5.3 Flash use?

1 message each on Pro, where it has 60 a day on Pro. The price of a message is fixed and shown before you send it, however long the reply.

How were these runs done?

The same way for every model: each prompt sent through llmwise's own pipeline, each reply checked the same way. The methods page has every prompt and how each is scored.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.