- Home
- LLM comparison
- GPT-5.4
Alternatives
What are the alternatives to GPT-5.4?
GPT-5.4 ranks 50 of 94 in the overall index and costs $5.63 per million tokens.
Claude Sonnet 5.5: Ahead in the overall index (121.8 vs. 86.5) and cheaper ($4 vs. $5.63 per million tokens).
Qwen 3.7 Plus: Level in the overall index (83.4 vs. 86.5) but cheaper: $0.48 instead of $5.63 per million tokens.
- Alternatives
- 5
- Overall index rank
- 50
- Data as of
- Nov 10, 2026
The alternatives by reason
Stronger, not much pricier
Claude Sonnet 5.5
Ahead in the overall index (121.8 vs. 86.5) and cheaper ($4 vs. $5.63 per million tokens).
- Overall index
- 121.8
- Price (3:1)
- $4
- Speed
- 103
Rank 5 · Upper group
Similar strength, cheaper
Qwen 3.7 Plus
Level in the overall index (83.4 vs. 86.5) but cheaper: $0.48 instead of $5.63 per million tokens.
- Overall index
- 83.4
- Price (3:1)
- $0.48
- Speed
- 50
Rank 54 · Lower group
With open weights
DeepSeek V4.1 Flash 1017
The best model with open weights, stronger: 106.7 vs. 86.5 in the overall index.
- Overall index
- 106.7
- Price (3:1)
- $0.26
- Speed
- 126
Rank 20 · Middle group
Head to head: GPT-5.4 vs. DeepSeek V4.1 Flash 1017Open the profile
Similar strength, faster
Gemini 3 Flash Preview
Level in the overall index (81.0 vs. 86.5) and faster: 101 instead of 58 tokens per second.
- Overall index
- 81.0
- Price (3:1)
- $1.13
- Speed
- 101
Rank 60 · Lower group
The successor
GPT-5.5
The successor to GPT-5.4. Overall index 101.5 vs. 86.5.
- Overall index
- 101.5
- Price (3:1)
- $11.25
- Speed
- 87
Rank 25 · Middle group
The models around GPT-5.4
| Rank | Model | Overall index | Price (3:1) | Speed |
|---|---|---|---|---|
| 47 | GLM-5.2 | 89.0 | $2.15 | 73 |
| 48 | DeepSeek-V4-Pro | 87.2 | $0.99 | 44 |
| 49 | Claude Sonnet 4.6 | 86.7 | $6 | 38 |
| 50 | GPT-5.4this model | 86.5 | $5.63 | 58 |
| 51 | Qwen 3.7 Max | 85.4 | $2.48 | 50 |
| 52 | Kimi K2.6 | 84.9 | $1.71 | 32 |
| 53 | Claude Opus 4.6 | 84.2 | $10 | 36 |
| 54 | Qwen 3.7 Plus | 83.4 | $0.48 | 50 |
Price in dollars per million tokens, mixed from 3 parts input and 1 part output. Speed in tokens per second.
Keep comparing
Profile of GPT-5.4Compare several models side by sideHow the overall index works