Claude Mythos Preview vs DeepSeek V4 Pro
Rank, tier, score and price side by side, from our weekly rankings. Scores are editorial opinion based on public sources.
A
Claude Mythos Preview
B
DeepSeek V4 Pro
- ProviderAnthropicDeepSeek
- Rank#33#30Higher rank
- TierAA
- Score82.483.0Higher score
- PriceInvite-only via Project Glasswing; $25 input / $125 output per 1M tokens after creditsAPI $1.32 input / $3.96 output per 1M tokens at peak hours ($0.66 / $1.98 off-peak); MIT open weightsLower price tier
- TaglineAnthropic's invite-only cybersecurity research preview offered through Project Glasswing, deprecated on the Claude API since June 2026.MIT-licensed open-weight heavyweight, among the open-weight leaders on SWE-bench Verified and LiveCodeBench.
| Attribute | Claude Mythos Preview | DeepSeek V4 Pro |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Rank | #33 | #30Higher rank |
| Tier | A | A |
| Score | 82.4 | 83.0Higher score |
| Price | Invite-only via Project Glasswing; $25 input / $125 output per 1M tokens after credits | API $1.32 input / $3.96 output per 1M tokens at peak hours ($0.66 / $1.98 off-peak); MIT open weightsLower price tier |
| Tagline | Anthropic's invite-only cybersecurity research preview offered through Project Glasswing, deprecated on the Claude API since June 2026. | MIT-licensed open-weight heavyweight, among the open-weight leaders on SWE-bench Verified and LiveCodeBench. |