Updated March 26, 2026· Based on independent benchmark data
GPT-5.3 Codex (xhigh) leads in intelligence with a score of 54.0 vs 10.5. LFM2 24B A2B is 58.3x cheaper at $0.03/1M tokens vs $1.75/1M. For speed, LFM2 24B A2B wins at 223 tok/s vs 72 tok/s.
| Metric | LFM2 24B A2B | GPT-5.3 Codex (xhigh) |
|---|---|---|
| Intelligence Score | 10.5 | 54.0 |
| Coding Score | 3.6 | 53.1 |
| Math Score | N/A | N/A |
| Speed (tok/s) | 223 tok/s | 72 tok/s |
| Latency (TTFT) | 0.23s | 83.91s |
| Input Price / 1M tokens | $0.03 | $1.75 |
| Output Price / 1M tokens | $0.12 | $14 |
| Context Window | N/A |
GPT-5.3 Codex (xhigh) outperforms LFM2 24B A2B on the intelligence index with a score of 54.0 compared to 10.5. For coding tasks, GPT-5.3 Codex (xhigh) has the edge with a coding score of 53.1 vs 3.6.
LFM2 24B A2B generates output significantly faster at 223 tok/s compared to GPT-5.3 Codex (xhigh)'s 72 tok/s, making it 3.1x faster for streaming responses. Time to first token is 0.23s for LFM2 24B A2B vs 83.91s for GPT-5.3 Codex (xhigh), which affects perceived responsiveness in interactive applications.
LFM2 24B A2B is more affordable at $0.03/1M input tokens ($0.12/1M output), while GPT-5.3 Codex (xhigh) costs $1.75/1M input ($14/1M output). That makes GPT-5.3 Codex (xhigh) 58.3x more expensive per token, which can add up significantly at scale. For a typical workload of 100 requests per day at 2,000 tokens each, LFM2 24B A2B would cost approximately $0.18/month vs $10.50/month for GPT-5.3 Codex (xhigh) in input costs alone.
Choose LFM2 24B A2B when you need faster output (223 tok/s), lower cost. Choose GPT-5.3 Codex (xhigh) when you need higher intelligence (54.0), stronger coding performance (53.1).
GPT-5.3 Codex (xhigh) scores higher on coding benchmarks (53.1 vs 3.6), making it the better choice for programming tasks.
LFM2 24B A2B is cheaper at $0.03/1M input tokens vs $1.75/1M for GPT-5.3 Codex (xhigh).
LFM2 24B A2B is faster, producing output at 223 tok/s compared to GPT-5.3 Codex (xhigh)'s 72 tok/s.
No, LFM2 24B A2B does not support image input. Neither model supports image input.
Data last synced: March 26, 2026
| N/A |
| Max Output Tokens | N/A | N/A |
| Input Modalities | Text | Text |
| Output Modalities | Text | Text |
| Free Tier | No | No |
It depends on your priorities. GPT-5.3 Codex (xhigh) scores higher on intelligence (54.0), but LFM2 24B A2B may be better for specific use cases like budget-conscious projects or speed-critical applications.