Cheaper overall
GLM 4.7 FlashX
45% lower blended rate
Higher benchmark scores
Nemotron-3.5 Super-120b-a12b
leads on 6 of 11 figures
Bigger context window
Nemotron-3.5 Super-120b-a12b
1,048,576 vs 202,752 tokens
Cheaper cached input
GLM 4.7 FlashX
$0.01 vs $0.04, 75% less
Where each one wins
Nemotron-3.5 Super-120b-a12b
- Larger context window 1,048,576 tokens
- Longer maximum output 262,144 tokens
- Higher overall intelligence (Intelligence Index) 26.5 vs 14.9
- Higher expert-exam performance (Humanity's Last Exam) 30.7 vs 7.6
- Higher long-context reasoning (AA-LCR) 69 vs 41.7
- Ahead on 3 more benchmark figures
GLM 4.7 FlashX
- Cheaper input tokens $0.07 vs $0.20, 65% less
- Cheaper output tokens $0.40 vs $0.50, 20% less
- Cheaper cached input $0.01 vs $0.04, 75% less
- Cheaper on all four workloads
- Higher 3d Elo 1,118 vs 1,117
- Higher Codecategories Elo 1,176 vs 1,153
- Higher Gamedev Elo 1,135 vs 1,112
- Ahead on 2 more benchmark figures
What four workloads cost
One run, and the same run a thousand times.
GLM 4.7 FlashX is cheaper on all four workloads, by much the same margin each time — about 2.2 times, from $0.0003 against $0.0005 on a chat turn to $0.0144 against $0.0390 on a long document.
| Workload | Nemotron-3.5 Super-120b-a12b | GLM 4.7 FlashX | ×1,000 | Difference |
|---|---|---|---|---|
| Chat turn 1,000 in / 500 out | $0.0005 | $0.0003 | $0.4500 / $0.2700 | GLM 4.7 FlashX is 40% cheaper |
| RAG answer 10,000 in / 800 out | $0.0024 | $0.0010 | $2.40 / $1.02 | GLM 4.7 FlashX is 58% cheaper |
| Agent loop 32,000 in / 700 out × 20 calls | $0.0390 | $0.0144 | $39.00 / $14.40 | GLM 4.7 FlashX is 63% cheaper |
| Long document 400,000 in / 3,000 out | GLM 4.7 FlashX's context window holds 202,752 tokens, so a 400,000-token prompt does not fit. | |||
Price per 1M tokens
Neither model changes its rate with prompt length, so these rates apply to every request.
| Context band | Price | Nemotron-3.5 Super-120b-a12b | GLM 4.7 FlashX | Difference |
|---|---|---|---|---|
| Any prompt length | Input | $0.20 | $0.07 | GLM 4.7 FlashX is 65% cheaper |
| Any prompt length | Cached input | $0.04 | $0.01 | GLM 4.7 FlashX is 75% cheaper |
| Any prompt length | Output | $0.50 | $0.40 | GLM 4.7 FlashX is 20% cheaper |
Blended: Nemotron-3.5 Super-120b-a12b $0.275, GLM 4.7 FlashX $0.1525 per 1M — one rate at 3:1 input to output, for comparing two models at a glance.
Nemotron-3.5 Super-120b-a12b priced by Perplexity, GLM 4.7 FlashX by Z.ai. Standard tier, pay-as-you-go.
Benchmarks
They share 11 figures: Nemotron-3.5 Super-120b-a12b leads on 6, GLM 4.7 FlashX on 5. The widest gap is 58.1 points, on answer reliability (Omniscience Non-Hallucination).
Overall intelligence
Intelligence Index — Overall intelligence across reasoning, knowledge & math evals
Nemotron-3.5 Super-120b-a12b ahead by 11.6
Expert-exam performance
Humanity's Last Exam — Expert-level questions across many academic domains
Nemotron-3.5 Super-120b-a12b ahead by 23.1
Long-context reasoning
AA-LCR — Long-context reasoning across large inputs
Nemotron-3.5 Super-120b-a12b ahead by 27.3
Physics research reasoning
CritPt — Unpublished physics research reasoning problems
Nemotron-3.5 Super-120b-a12b ahead by 10.3
Factual accuracy
Omniscience Accuracy — Breadth of factual knowledge across domains
Nemotron-3.5 Super-120b-a12b ahead by 7.1
Answer reliability
Omniscience Non-Hallucination — How reliably the model avoids fabricated answers
Nemotron-3.5 Super-120b-a12b ahead by 58.1
3d Elo
Design Arena head-to-head rating in the 3d category
GLM 4.7 FlashX ahead by 1
Codecategories Elo
Design Arena head-to-head rating in the codecategories category
GLM 4.7 FlashX ahead by 23
Gamedev Elo
Design Arena head-to-head rating in the gamedev category
GLM 4.7 FlashX ahead by 23
Uicomponent Elo
Design Arena head-to-head rating in the uicomponent category
GLM 4.7 FlashX ahead by 53
Website Elo
Design Arena head-to-head rating in the website category
GLM 4.7 FlashX ahead by 23
Each figure is that model's best published run, with the effort level named beside it — where these scores come from.
Specifications
Nemotron-3.5 Super-120b-a12b holds 845,824 more input tokens in one request. Nemotron-3.5 Super-120b-a12b also takes audio, file, image and video. Only Nemotron-3.5 Super-120b-a12b lists code execution; only GLM 4.7 FlashX lists prompt caching.
| Specification | Nemotron-3.5 Super-120b-a12b | GLM 4.7 FlashX |
|---|---|---|
| Context window | 1,048,576 tokens | 202,752 tokens |
| Max output | 262,144 tokens | 128,000 tokens |
| Takes in | Text, audio, image, video, file | Text |
| Puts out | Text | Text |
| Released | 15 Oct 2026 | 19 Jan 2026 |
| Status | Active | Active |
| Sold by | Perplexity | Z.ai |
| Tool use | Yes | Yes |
| Structured outputs | Yes | Yes |
| Web search | Not listed | Not listed |
| Prompt caching | Not listed | Yes |
| Code execution | Yes | Not listed |
| Computer use | Not listed | Not listed |
| Open weights | Yes | Yes |
| Reasoning | Yes | Yes |
FAQs
- Is Nemotron-3.5 Super-120b-a12b cheaper than GLM 4.7 FlashX?
- No — the other way round. A 1,000-token prompt with a 500-token reply costs $0.0005 on Nemotron-3.5 Super-120b-a12b and $0.0003 on GLM 4.7 FlashX, and the same model is cheaper on every workload on this page.
- How much do Nemotron-3.5 Super-120b-a12b and GLM 4.7 FlashX cost per 1M tokens?
- Nemotron-3.5 Super-120b-a12b costs $0.20 for input and $0.50 for output. GLM 4.7 FlashX costs $0.07 and $0.40. Standard pay-as-you-go rates, as of 16 Oct 2026.
- Which is better, Nemotron-3.5 Super-120b-a12b or GLM 4.7 FlashX?
- On published benchmarks, Nemotron-3.5 Super-120b-a12b. It leads on 6 of the 11 figures both models report, including overall intelligence (Intelligence Index), where it scores 26.5 against 14.9.
- Does Nemotron-3.5 Super-120b-a12b or GLM 4.7 FlashX have a bigger context window?
- Nemotron-3.5 Super-120b-a12b, at 1,048,576 tokens against 202,752. That is 845,824 more input tokens in a single request.
Explore these models
Perplexity · 1,048,576-token context