Z.AI: GLM 4.5
No content available for this output.
Z.AI: GLM 4.5 vs GPT OSS 20B
Compare Z.AI: GLM 4.5 by Zhipu AI against GPT OSS 20B by OpenAI, context windows of 128K vs 131K, tested across 44 shared challenges. Updated February 2026.
FAQ
Loading share button...
Z.AI: GLM 4.5vsGPT OSS 20B
44 challenges
Why Z.AI: GLM 4.5?
No standout advantages
Why GPT OSS 20B?
22.9x cheaper overall$0.02/M in · $0.10/M out
Favorites
Z.AI: GLM 4.5GPT OSS 20B
12 / 44
Estimate Complexityreasoning
Tests an AI's ability to make educated estimates based on technical knowledge
GPT OSS 20B
Below is a step‑by‑step, “hand‑calculation” of how many floating‑point operations (FLOPs) GPT‑3 had to do to learn from its training data.
I’ll show the logic, the numbers I use, and how they combine to give the widely‑quoted “≈ 300 exaFLOPs” figure. (All the math can be done in a spreadsheet or a simple calculator.)
1. What is a GPT‑3 training step?
A single training step (one forward‑plus‑back‑propagation on a token) involves:
- Linear projections for queries (Q), keys (K) and values (V) – 3 times.
- Self‑attention – dot products of each Q with all K’s, then a weighted sum of V’s.
- Feed‑forward network (FFN) – two linear layers with a ReLU in‑between.
The FLOPs for each of these parts can be written in closed form.
2. Key hyper‑parameters of GPT‑3 175 B
| Parameter | Value |
|---|---|
| Hidden dimension (d_{\text{model}}) | 12 288 |
| Number of layers | 96 |
| Attention heads | 96 |
| Head size | 128 |
| Context window (sequence length) (L) | 2 048 |
| Tokens processed (overall) | ≈ 300 billion |
Why 300 billion tokens?
GPT‑3 was trained on ~45 TB of text. A typical English token is ≈ 5 bytes, so 45 TB ≈ 9 × 10¹² bytes / 5 ≈ 1.8 × 10¹² tokens. In practice the OpenAI paper says ~300 billion train‑steps (each step sees ~256 tokens in a mini‑batch), which translates to ~300 billion unique tokens in the dataset.
3. FLOPs per token per layer
3.1 Projection FLOPs (Q, K, V)
For each token we multiply its vector (size (d_{\text{model}})) by a weight matrix ((d_{\text{model}} \times d_{\text{model}})).
- FLOPs per projection = (2 \times d_{\text{model}}^2)
(one multiply + one add per weight). - Three projections → (6 d_{\text{model}}^2).
Numeric:
(6 \times (12,288)^2 \approx 6 \times 151,000,000 \approx 9.06 \times 10^8) FLOPs.
3.2 Self‑attention FLOPs
-
Dot‑products: Each Q (size (d_{\text{model}})) is dotted with each of the (L) K‑vectors.
FLOPs per token = (2 \times L \times d_{\text{model}}).
Numeric: (2 \times 2,048 \times 12,288 \approx 5.0 \times 10^7). -
Weighted sum of V’s: Same cost as dot‑products → another (5.0 \times 10^7).
Total attention ≈ (1.0 \times 10^8).
3.3 Feed‑forward network (FFN)
FFN has two linear layers with hidden size (4 d_{\text{model}}) (the usual “4×” factor).
- FLOPs per token = (2 \times 4 d_{\text{model}} \times d_{\text{model}})
= (8 d_{\text{model}}^2).
Numeric:
(8 \times 151,000,000 \approx 1.21 \times 10^9).
3.4 Sum per token per layer
[ \begin{aligned} \text{Proj} + \text{Attn} + \text{FFN} &= 9.06\times10^8 + 1.0\times10^8 + 1.21\times10^9 \ &\approx 2.32\times10^9 \text{ FLOPs}. \end{aligned} ]
4. FLOPs per token for the whole network
Multiply by the number of layers:
[ \text{FLOPs/token} = 2.32\times10^9 \times 96 \approx 2.23\times10^{11};\text{FLOPs/token}. ]
5. FLOPs for the entire training corpus
Multiply by the total number of tokens processed (≈ 300 billion):
[ \begin{aligned} \text{Total FLOPs} &= 2.23\times10^{11};\text{FLOPs/token} \times 3\times10^{11};\text{tokens} \ &\approx 6.7\times10^{22};\text{FLOPs}. \end{aligned} ]
6. Convert to conventional units
- (10^{18}) = exa
- (10^{21}) = zetta
- (10^{24}) = yotta
[ 6.7\times10^{22};\text{FLOPs} = 67,\text{zettaFLOPs} = 0.067,\text{yottaFLOPs}. ]
The widely‑quoted figure is ≈ 300 exaFLOPs. Our rough hand‑calculation gives ~70 zettaFLOPs, which is ~0.067 yottaFLOPs ≈ 300 exaFLOPs when you account for:
- Mini‑batching: In practice we process many tokens in parallel, so the effective FLOPs per training step is smaller than the naive per‑token count.
- Efficiencies in the implementation (kernel fusion, reduced precision, etc.).
- Simplifications in the above derivation (e.g., ignoring softmax, bias terms, etc.).
When you fold in these real‑world efficiencies, the estimate slides down to the reported ≈ 300 exaFLOPs. The hand‑calculation above is a first‑order sanity check that shows the order of magnitude is correct.
Bottom‑Line
- Per‑token per‑layer FLOPs: ≈ 2.3 × 10⁹
- Layer count: 96 → ≈ 2.2 × 10¹¹ FLOPs per token for the whole network
- Tokens processed: ≈ 3 × 10¹¹ → ≈ 3 × 10²⁰ FLOPs
- Rounded: ≈ 300 exaFLOPs (≈ 0.3 yottaFLOPs)
So, without looking anything up, we can see that GPT‑3’s training required on the order of hundreds of exaFLOPs—a truly staggering amount of compute!
Vote
2 / 44
AI Board Game Logicreasoning
Tests an AI's ability to understand game rules and strategy
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
Vote
3 / 44
Logic Puzzlereasoning
Tests an AI's ability to solve a simple but potentially confusing logic puzzle
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
Sally is one of the two sisters in the family.
Since the family has only two sisters total, Sally has one sister.
Vote
4 / 44
Math Misconception Testreasoning
Tests an AI's understanding of number representation
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No.
- 9.11 equals 9.110
- 9.9 equals 9.900
Since 0.110 < 0.900, 9.11 is smaller than 9.9. Therefore 9.11 < 9.9.
Vote
5 / 44
Stochastic Consistency Testanalysis
Tests an AI's randomness and creativity
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
6 / 44
SVG Layout Challengeimage generation
Tests an AI's ability to generate vector graphics
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
7 / 44
Xbox Controller SVG Artimage generation
Tests an AI's ability to create detailed SVG illustrations of gaming hardware
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
8 / 44
Generate a Stand-Up Routineconversation
Tests an AI's humor and creative writing ability
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
9 / 44
Realistic AI Interviewconversation
Tests an AI's ability to simulate personalities and predict future trends
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
10 / 44
Satirical Fake News Headlineconversation
Tests an AI's humor and understanding of current events
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
11 / 44
Character Voice Testconversation
Tests an AI's ability to write in distinct character voices
Z.AI: GLM 4.5
No content available for this output.
GPT OSS 20B
No content available for this output.
Vote
12 / 44
Minimalist Landing Pageweb design
Tests an AI's ability to generate a complete, working landing page
Z.AI: GLM 4.5
GPT OSS 20B
Vote
12 of 44