Baidu: ERNIE-Light-DT-1.0 vs Claude Opus 5 5 — API Pricing, 150K Cache & Latency Benchmark
Side-by-side unit economics audit comparing Baidu: ERNIE-Light-DT-1.0 (baidu/ernie-light-dt-1.0-intl: Input $1.13/1M, Output $3.94/1M, Warm Cache Read $0.113/1M) against Claude Opus 5 5 (claude-opus-5-5: Input $4/1M, Output $20/1M, Warm Cache Read $0.2/1M).
Activate 4-breakpoint prefix caching on Baidu: ERNIE-Light-DT-1.0 to cut input token costs from $1.13/1M down to $0.113/1M, or route async workloads via Batch API for 50% off.
Lock static system prompt prefixes on Claude Opus 5 5 to pay $0.2/1M warm cache rates instead of $4/1M uncached input, plus eligible startup cloud credits.
In Etuki FinOps telemetry comparing Baidu: ERNIE-Light-DT-1.0 (API ID: baidu/ernie-light-dt-1.0-intl) vs Claude Opus 5 5 (API ID: claude-opus-5-5), Baidu: ERNIE-Light-DT-1.0 offers lower standard uncached input pricing at $1.13/1M tokens, while Baidu: ERNIE-Light-DT-1.0 leads in 10-turn 150K cache-pinned coding economics at $0.4008 per 10 turns (saving $1.0192 over Claude Opus 5 5). For latency-sensitive interactive agents, Claude Opus 5 5 delivers 12ms faster P90 time-to-first-token (420ms vs 432ms).
| Specification / FinOps Dimension | Baidu: ERNIE-Light-DT-1.0 (baidu/ernie-light-dt-1.0-intl) | Claude Opus 5 5 (claude-opus-5-5) | Verified Advantage |
|---|---|---|---|
| Official Vendor API ID | baidu/ernie-light-dt-1.0-intl | claude-opus-5-5 | UMCE 563-Registry Locked |
| 563-Registry & Routing Eligibility | is_known: true | routable: true | is_known: true | routable: true | Both Merkle Verified |
| Standard Uncached Input (/1M tokens) | $1.13 / 1M | $4 / 1M | Baidu: ERNIE-Light-DT-1.0 |
| Warm Prefix Cache Read (/1M tokens) | $0.113 / 1M | $0.2 / 1M | Baidu: ERNIE-Light-DT-1.0 |
| Standard Output (/1M tokens) | $3.94 / 1M | $20 / 1M | Baidu: ERNIE-Light-DT-1.0 |
| Context Window Capacity | 1024K Tokens | 200K Tokens | Baidu: ERNIE-Light-DT-1.0 |
| 150K Prompt × 10 Turns (Unpinned Cold + 2K Out) | $1.7738 | $6.4000 | Baidu: ERNIE-Light-DT-1.0 |
| 150K × 10 Turns Case A (100% Static Pinned) | $0.4008 (saves $1.3730) | $1.4200 (saves $4.9800) | Baidu: ERNIE-Light-DT-1.0 (-$1.0192) |
| 150K × 10 Turns Case B (90% Static + 10% Tail) | $0.5381 (saves $1.2357) | $1.9330 (saves $4.4670) | Baidu: ERNIE-Light-DT-1.0 |
| P50 / P90 First-Token Latency | 294ms / 432ms (89 tok/s) | 286ms / 420ms (88 tok/s) | Claude Opus 5 5 (-12ms) |
1. Full-Turn Rate-Card Substitution for Baidu: ERNIE-Light-DT-1.0: At an uncached input rate of $1.13/1M, output rate of $3.94/1M, and warm cache read rate of $0.113/1M, running 10 consecutive 150,000-token coding agent turns (with 2,000 output tokens/turn) without prefix pinning costs $1.7738. Locking 100% static prefixes (Case A: 1 cold turn + 9 warm cached turns + 20K total output) reduces 10-turn spend to $0.4008 (saving $1.3730); under a 90% static prefix + 10% dynamic tail workload (Case B), 10-turn spend is $0.5381.
2. Full-Turn Rate-Card Substitution for Claude Opus 5 5 & Winner Verdict: On Claude Opus 5 5 ($4/1M input, $20/1M output, $0.2/1M warm cache), 10 unpinned 150K turns cost $6.4000, dropping to $1.4200 in Case A and $1.9330 in Case B. Across every 10-turn 150K coding loop (including output tokens), Baidu: ERNIE-Light-DT-1.0 delivers an additional $1.0192 Case-A cost advantage.
Production Workload TCO: 25,000 MAU • 50M Tokens/Mo (50% Cache Hit) • 2,000 GB Egress
💡 Optimal FinOps Verdict: Deploying Baidu: ERNIE-Light-DT-1.0 saves $219.74/month ($2637/year) at 25,000 MAU & 50M monthly tokens (80% input / 20% output split with 50% prefix cache hit rate).