Cheaper overall
GLM 5.3
46% lower blended rate
Higher benchmark scores
GPT-6 Sol
leads on 5 of 8 figures
Bigger context window
GPT-6 Sol
1,050,000 vs 1,000,000 tokens
Cheaper cached input
GPT-6 Sol
$0.20 vs $0.26, 23% less
Where each one wins
GPT-6 Sol
- Cheaper cached input $0.20 vs $0.26, 23% less
- Larger context window 1,050,000 tokens
- Higher overall intelligence (Intelligence Index) 47.5 vs 44.8
- Higher expert-exam performance (Humanity's Last Exam) 47.9 vs 42.3
- Higher long-context reasoning (AA-LCR) 83.7 vs 79.7
- Ahead on 2 more benchmark figures
GLM 5.3
- Cheaper input tokens $1.40 vs $2.00, 30% less
- Cheaper output tokens $4.40 vs $10.00, 56% less
- Cheaper on all four workloads
- Higher scientific coding (SciCode) 59 vs 57.6
- Higher real-world task performance (GDPval) 57.2 vs 49.3
- Higher answer reliability (Omniscience Non-Hallucination) 70.4 vs 39.9
What four workloads cost
One run, and the same run a thousand times.
The gap opens on long prompts: GLM 5.3 costs $0.5732 against $1.65, a difference of $1,071.80 over a thousand runs.
| Workload | GPT-6 Sol | GLM 5.3 | ×1,000 | Difference |
|---|---|---|---|---|
| Chat turn 1,000 in / 500 out | $0.0070 | $0.0036 | $7.00 / $3.60 | GLM 5.3 is 49% cheaper |
| RAG answer 10,000 in / 800 out | $0.0280 | $0.0175 | $28.00 / $17.52 | GLM 5.3 is 37% cheaper |
| Agent loop 32,000 in / 700 out × 20 calls | $0.3400 | $0.2736 | $340.00 / $273.60 | GLM 5.3 is 20% cheaper |
| Long document 400,000 in / 3,000 out | $1.65 | $0.5732 | $1,645.00 / $573.20 | GLM 5.3 is 65% cheaper |
Price per 1M tokens
Above 272,000 input tokens the rates change, and the new rate applies to the whole request, not just the tokens past that point: GPT-6 Sol's input goes from $2.00 to $4.00. It does not change which of the two is cheaper.
| Context band | Price | GPT-6 Sol | GLM 5.3 | Difference |
|---|---|---|---|---|
| Prompts up to 272,000 tokens | Input | $2.00 | $1.40 | GLM 5.3 is 30% cheaper |
| Prompts up to 272,000 tokens | Cached input | $0.20 | $0.26 | GPT-6 Sol is 23% cheaper |
| Prompts up to 272,000 tokens | Output | $10.00 | $4.40 | GLM 5.3 is 56% cheaper |
| Prompts over 272,000 tokens | Input | $4.00 | $1.40 | GLM 5.3 is 65% cheaper |
| Prompts over 272,000 tokens | Cached input | $0.40 | $0.26 | GLM 5.3 is 35% cheaper |
| Prompts over 272,000 tokens | Output | $15.00 | $4.40 | GLM 5.3 is 71% cheaper |
Blended: GPT-6 Sol $4.00, GLM 5.3 $2.15 per 1M — one rate at 3:1 input to output, for comparing two models at a glance.
GPT-6 Sol priced by Perplexity, GLM 5.3 by Z.ai. Standard tier, pay-as-you-go.
Benchmarks
They share 8 figures: GPT-6 Sol leads on 5, GLM 5.3 on 3. The widest gap is 30.5 points, on answer reliability (Omniscience Non-Hallucination).
Overall intelligence
Intelligence Index — Overall intelligence across reasoning, knowledge & math evals
GPT-6 Sol ahead by 2.7
Expert-exam performance
Humanity's Last Exam — Expert-level questions across many academic domains
GPT-6 Sol ahead by 5.6
Long-context reasoning
AA-LCR — Long-context reasoning across large inputs
GPT-6 Sol ahead by 4
Scientific coding
SciCode — Research-level scientific coding tasks
GLM 5.3 ahead by 1.4
Physics research reasoning
CritPt — Unpublished physics research reasoning problems
GPT-6 Sol ahead by 11.8
Real-world task performance
GDPval — Economically valuable, real-world knowledge work
GLM 5.3 ahead by 7.9
Factual accuracy
Omniscience Accuracy — Breadth of factual knowledge across domains
GPT-6 Sol ahead by 20.6
Answer reliability
Omniscience Non-Hallucination — How reliably the model avoids fabricated answers
GLM 5.3 ahead by 30.5
Each figure is that model's best published run, with the effort level named beside it — where these scores come from.
Specifications
GPT-6 Sol holds 50,000 more input tokens in one request. GPT-6 Sol also takes file and image. Only GPT-6 Sol lists computer use; only GLM 5.3 lists open weights.
| Specification | GPT-6 Sol | GLM 5.3 |
|---|---|---|
| Context window | 1,050,000 tokens | 1,000,000 tokens |
| Max output | 128,000 tokens | 128,000 tokens |
| Takes in | Text, image, file | Text |
| Puts out | Text | Text |
| Knowledge cutoff | 20 Apr 2026 | — |
| Released | 22 Sep 2026 | 14 Aug 2026 |
| Status | Active | Active |
| Sold by | Perplexity | Perplexity, Z.ai |
| Tool use | Yes | Yes |
| Structured outputs | Yes | Yes |
| Web search | Not listed | Not listed |
| Prompt caching | Yes | Yes |
| Code execution | Yes | Yes |
| Computer use | Yes | Not listed |
| Open weights | Not listed | Yes |
| Reasoning | Yes | Yes |
FAQs
- Is GPT-6 Sol cheaper than GLM 5.3?
- No — the other way round. A 1,000-token prompt with a 500-token reply costs $0.0070 on GPT-6 Sol and $0.0036 on GLM 5.3, and the same model is cheaper on every workload on this page.
- How much do GPT-6 Sol and GLM 5.3 cost per 1M tokens?
- GPT-6 Sol costs $2.00 for input and $10.00 for output. GLM 5.3 costs $1.40 and $4.40. Standard pay-as-you-go rates, as of 23 Sep 2026.
- Which is better, GPT-6 Sol or GLM 5.3?
- On published benchmarks, GPT-6 Sol. It leads on 5 of the 8 figures both models report, including overall intelligence (Intelligence Index), where it scores 47.5 against 44.8.
- Does GPT-6 Sol or GLM 5.3 have a bigger context window?
- GPT-6 Sol, at 1,050,000 tokens against 1,000,000. That is 50,000 more input tokens in a single request.