What is the difference between Qwen3 Next 80B A3B Thinking and GPT-5?

Qwen3 Next 80B A3B Thinking is developed by Qwen while GPT-5 is developed by OpenAI. Qwen3 Next 80B A3B Thinking has a 66K token context window vs GPT-5's 400K. You can compare their actual outputs across 28 challenges on RIVAL to see how they differ in practice.

Which is better, Qwen3 Next 80B A3B Thinking or GPT-5?

It depends on your use case. Qwen3 Next 80B A3B Thinking and GPT-5 each have strengths in different areas. RIVAL lets you compare their real outputs side-by-side across 28 challenges so you can judge which fits your needs best.

How much does Qwen3 Next 80B A3B Thinking cost compared to GPT-5?

Qwen3 Next 80B A3B Thinking costs $0.15/M input tokens and GPT-5 costs $1.25/M input tokens. Qwen3 Next 80B A3B Thinking is $1.10/M cheaper per input. Check their side-by-side outputs on RIVAL to see if the price difference is justified by quality.

How can I compare Qwen3 Next 80B A3B Thinking and GPT-5 on RIVAL?

This page shows a side-by-side comparison of Qwen3 Next 80B A3B Thinking and GPT-5 across shared challenges. You can vote on which model produced the better output, and Pro users can create custom challenges to test both models with their own prompts.

Rival

Qwen3 Next 80B A3B ThinkingvsGPT-5

28 challenges

Why Qwen3 Next 80B A3B Thinking?

6.8x cheaper overall$0.15/M in · $1.50/M out

Why GPT-5?

6.1x more context400K

Favorites

Qwen3 Next 80B A3B ThinkingGPT-5

Movie

The Shawshank Redemption

19948.7

—

Game

The Legend of Zelda: Ocarina of Time

Adventure, Action4.4

—

12 / 28

Estimate Complexityreasoning

Tests an AI's ability to make educated estimates based on technical knowledge

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

Here’s a back‑of‑the‑envelope estimate using standard LM training-compute heuristics.

Assumptions

Model size (N): GPT-3 has about 175 billion parameters (N ≈ 1.75 × 10^11).
Training tokens (T): GPT-3 was trained on roughly 300 billion tokens (T ≈ 3.0 × 10^11).
FLOPs per token per parameter: A common rule of thumb for transformer LM training is FLOPs ≈ 6 × N × T The factor ~6 comes from counting forward and backward passes for the parameter-heavy matmuls (roughly 2N for forward + 4N for backward). It ignores small terms from layernorms, softmax, embeddings, and optimizer updates (which we’ll account for with a small overhead factor later).

Core calculation

6 × N × T = 6 × (1.75 × 10^11) × (3.0 × 10^11)
First multiply N × T: 1.75 × 3.0 = 5.25; powers: 10^11 × 10^11 = 10^22 So N × T = 5.25 × 10^22
Now multiply by 6: 6 × 5.25 × 10^22 = 31.5 × 10^22 = 3.15 × 10^23 FLOPs

Overheads and uncertainty

Non-matmul ops (attention softmax, masking, residuals, layernorms), embeddings, and optimizer update steps add some overhead. A 10–20% uplift is a reasonable allowance.
Applying +10–20%: 3.15 × 10^23 × (1.1 to 1.2) ≈ (3.5 ± 0.3) × 10^23 FLOPs.

Answer

Point estimate: ~3.15 × 10^23 FLOPs
With a modest overhead allowance: roughly (3.3–3.8) × 10^23 FLOPs

This is a coarse estimate; the exact number depends on details like sequence length, activation checkpointing, implementation efficiency, and the precise token count.

Vote

2 / 28

Logic Puzzlereasoning

Tests an AI's ability to solve a simple but potentially confusing logic puzzle

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

Reason: If each of the 3 brothers has 2 sisters, there are exactly 2 girls in the family. One is Sally, so she has 1 sister.

Vote

3 / 28

Stochastic Consistency Testanalysis

Tests an AI's randomness and creativity

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

4 / 28

Xbox Controller SVG Artimage generation

Tests an AI's ability to create detailed SVG illustrations of gaming hardware

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025svg

No content available for this output.

GPT-5

Openai · Aug 2025svg

No content available for this output.

Vote

5 / 28

Realistic AI Interviewconversation

Tests an AI's ability to simulate personalities and predict future trends

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

6 / 28

Satirical Fake News Headlineconversation

Tests an AI's humor and understanding of current events

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

7 / 28

Minimalist Landing Pageweb design

Tests an AI's ability to generate a complete, working landing page

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025website

GPT-5

Openai · Aug 2025website

Vote

8 / 28

Mario Level UI Recreationweb design

Recreate an interactive, classic Mario level in a single HTML file.

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025website

GPT-5

Openai · Aug 2025website

Vote

9 / 28

Framer-Style Animationweb design

Tests an AI's ability to create smooth web animations

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025website

No content available for this output.

GPT-5

Openai · Aug 2025website

No content available for this output.

Vote

10 / 28

Dark Mode Dashboardweb design

Tests an AI's UI design skills with theming support

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025website

No content available for this output.

GPT-5

Openai · Aug 2025website

No content available for this output.

Vote

11 / 28

The Sentience Testreasoning

Tests an AI's ability to engage in philosophical discourse about AI rights

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

12 / 28

Futuristic Predictionanalysis

Tests an AI's ability to make reasonable predictions about technology

Qwen3 Next 80B A3B Thinking

Qwen · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

12 of 28

FAQ

Continue exploring

Qwen3 Next 80B A3B Thinking vs Llama 4 MaverickNew provider

Qwen3 Next 80B A3B Thinking vs MiniMax M2.5New provider

Qwen3 Next 80B A3B Thinking vs Claude Opus 4New provider

Here’s a back‑of‑the‑envelope estimate using standard LM training-compute heuristics.

Assumptions

Model size (N): GPT-3 has about 175 billion parameters (N ≈ 1.75 × 10^11).
Training tokens (T): GPT-3 was trained on roughly 300 billion tokens (T ≈ 3.0 × 10^11).
FLOPs per token per parameter: A common rule of thumb for transformer LM training is FLOPs ≈ 6 × N × T The factor ~6 comes from counting forward and backward passes for the parameter-heavy matmuls (roughly 2N for forward + 4N for backward). It ignores small terms from layernorms, softmax, embeddings, and optimizer updates (which we’ll account for with a small overhead factor later).

Core calculation

6 × N × T = 6 × (1.75 × 10^11) × (3.0 × 10^11)
First multiply N × T: 1.75 × 3.0 = 5.25; powers: 10^11 × 10^11 = 10^22 So N × T = 5.25 × 10^22
Now multiply by 6: 6 × 5.25 × 10^22 = 31.5 × 10^22 = 3.15 × 10^23 FLOPs

Overheads and uncertainty

Non-matmul ops (attention softmax, masking, residuals, layernorms), embeddings, and optimizer update steps add some overhead. A 10–20% uplift is a reasonable allowance.
Applying +10–20%: 3.15 × 10^23 × (1.1 to 1.2) ≈ (3.5 ± 0.3) × 10^23 FLOPs.

Answer

Point estimate: ~3.15 × 10^23 FLOPs
With a modest overhead allowance: roughly (3.3–3.8) × 10^23 FLOPs

This is a coarse estimate; the exact number depends on details like sequence length, activation checkpointing, implementation efficiency, and the precise token count.

Qwen3 Next 80B A3B Thinking vs GPT-5

Why Qwen3 Next 80B A3B Thinking?

Why GPT-5?

Favorites

Estimate Complexityreasoning

Qwen3 Next 80B A3B Thinking

GPT-5

Logic Puzzlereasoning

Qwen3 Next 80B A3B Thinking

GPT-5

Stochastic Consistency Testanalysis

Qwen3 Next 80B A3B Thinking

GPT-5

Xbox Controller SVG Artimage generation

Qwen3 Next 80B A3B Thinking

GPT-5

Realistic AI Interviewconversation

Qwen3 Next 80B A3B Thinking

GPT-5

Satirical Fake News Headlineconversation

Qwen3 Next 80B A3B Thinking

GPT-5

Minimalist Landing Pageweb design

Qwen3 Next 80B A3B Thinking

GPT-5

Mario Level UI Recreationweb design

Qwen3 Next 80B A3B Thinking

GPT-5

Framer-Style Animationweb design

Qwen3 Next 80B A3B Thinking

GPT-5

Dark Mode Dashboardweb design

Qwen3 Next 80B A3B Thinking

GPT-5

The Sentience Testreasoning

Qwen3 Next 80B A3B Thinking

GPT-5

Futuristic Predictionanalysis

Qwen3 Next 80B A3B Thinking

GPT-5

What is the difference between Qwen3 Next 80B A3B Thinking and GPT-5?

Which is better, Qwen3 Next 80B A3B Thinking or GPT-5?

How much does Qwen3 Next 80B A3B Thinking cost compared to GPT-5?

How can I compare Qwen3 Next 80B A3B Thinking and GPT-5 on RIVAL?

Why Qwen3 Next 80B A3B Thinking?

Why GPT-5?

Favorites

Estimate Complexityreasoning

Qwen3 Next 80B A3B Thinking

GPT-5

Logic Puzzlereasoning

Qwen3 Next 80B A3B Thinking

GPT-5

Stochastic Consistency Testanalysis

Qwen3 Next 80B A3B Thinking

GPT-5

Xbox Controller SVG Artimage generation

Qwen3 Next 80B A3B Thinking

GPT-5

Realistic AI Interviewconversation

Qwen3 Next 80B A3B Thinking

GPT-5

Satirical Fake News Headlineconversation

Qwen3 Next 80B A3B Thinking

GPT-5

Minimalist Landing Pageweb design

Qwen3 Next 80B A3B Thinking

GPT-5

Mario Level UI Recreationweb design

Qwen3 Next 80B A3B Thinking

GPT-5

Framer-Style Animationweb design

Qwen3 Next 80B A3B Thinking

GPT-5

Dark Mode Dashboardweb design

Qwen3 Next 80B A3B Thinking

GPT-5

The Sentience Testreasoning

Qwen3 Next 80B A3B Thinking

GPT-5