1 vs 0 benchmarks won
DeepSeek DeepSeek-V4-Flash | OpenAI GPT-5.3-Codex-Spark | |
|---|---|---|
| Overview | ||
| Company | DeepSeek | OpenAI |
| Release date | Apr 24 2026 | Feb 12 2026 |
| Model type | — | — |
| Open source | Yes | No |
| Specifications | ||
Parameters | — | — |
Context window | — | 128k |
| Benchmarks | ||
Science reasoning GPQA Diamond | 88.1% | — |
Software engineering SWE-Bench Verified | — | — |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | GPT-5.3-Codex-Spark shipped 71 days before DeepSeek-V4-Flash | |
DeepSeek-V4-Flash leads GPT-5.3-Codex-Spark on 1 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). GPT-5.3-Codex-Spark shipped 71 days before DeepSeek-V4-Flash, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Flash is an open-source / open-weight model; GPT-5.3-Codex-Spark is proprietary.
Direct benchmark comparisons are unavailable — at least one of these models has not published scores on GPQA Diamond, SWE-Bench Verified, or MMMU.