Claude Sonnet 5 5 vs Thinking Machines: Inkling 2 Small — API Pricing, 150K Cache & Latency Benchmark
Side-by-side unit economics audit comparing Claude Sonnet 5 5 (claude-sonnet-5-5: Input $2/1M, Output $10/1M, Warm Cache Read $0.1/1M) against Thinking Machines: Inkling 2 Small (thinkingmachines/inkling-2-small: Input $0.3/1M, Output $1/1M, Warm Cache Read $0.03/1M).
Activate 4-breakpoint prefix caching on Claude Sonnet 5 5 to cut input token costs from $2/1M down to $0.1/1M, or route async workloads via Batch API for 50% off.
Lock static system prompt prefixes on Thinking Machines: Inkling 2 Small to pay $0.03/1M warm cache rates instead of $0.3/1M uncached input, plus eligible startup cloud credits.
In Etuki FinOps telemetry comparing Claude Sonnet 5 5 (API ID: claude-sonnet-5-5) vs Thinking Machines: Inkling 2 Small (API ID: thinkingmachines/inkling-2-small), Thinking Machines: Inkling 2 Small offers lower standard uncached input pricing at $0.3/1M tokens, while Thinking Machines: Inkling 2 Small leads in 10-turn 150K cache-pinned coding economics at $0.1055 per 10 turns (saving $0.6045 over Claude Sonnet 5 5). For latency-sensitive interactive agents, Claude Sonnet 5 5 delivers 20ms faster P90 time-to-first-token (310ms vs 330ms).
| Specification / FinOps Dimension | Claude Sonnet 5 5 (claude-sonnet-5-5) | Thinking Machines: Inkling 2 Small (thinkingmachines/inkling-2-small) | Verified Advantage |
|---|---|---|---|
| Official Vendor API ID | claude-sonnet-5-5 | thinkingmachines/inkling-2-small | 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) | $2 / 1M | $0.3 / 1M | Thinking Machines: Inkling 2 Small |
| Warm Prefix Cache Read (/1M tokens) | $0.1 / 1M | $0.03 / 1M | Thinking Machines: Inkling 2 Small |
| Standard Output (/1M tokens) | $10 / 1M | $1 / 1M | Thinking Machines: Inkling 2 Small |
| Context Window Capacity | 200K Tokens | 1049K Tokens | Thinking Machines: Inkling 2 Small |
| 150K Prompt × 10 Turns (Unpinned Cold + 2K Out) | $3.2000 | $0.4700 | Thinking Machines: Inkling 2 Small |
| 150K × 10 Turns Case A (100% Static Pinned) | $0.7100 (saves $2.4900) | $0.1055 (saves $0.3645) | Thinking Machines: Inkling 2 Small (-$0.6045) |
| 150K × 10 Turns Case B (90% Static + 10% Tail) | $0.9665 (saves $2.2335) | $0.1419 (saves $0.3281) | Thinking Machines: Inkling 2 Small |
| P50 / P90 First-Token Latency | 211ms / 310ms (108 tok/s) | 224ms / 330ms (112 tok/s) | Claude Sonnet 5 5 (-20ms) |
1. Full-Turn Rate-Card Substitution for Claude Sonnet 5 5: At an uncached input rate of $2/1M, output rate of $10/1M, and warm cache read rate of $0.1/1M, running 10 consecutive 150,000-token coding agent turns (with 2,000 output tokens/turn) without prefix pinning costs $3.2000. Locking 100% static prefixes (Case A: 1 cold turn + 9 warm cached turns + 20K total output) reduces 10-turn spend to $0.7100 (saving $2.4900); under a 90% static prefix + 10% dynamic tail workload (Case B), 10-turn spend is $0.9665.
2. Full-Turn Rate-Card Substitution for Thinking Machines: Inkling 2 Small & Winner Verdict: On Thinking Machines: Inkling 2 Small ($0.3/1M input, $1/1M output, $0.03/1M warm cache), 10 unpinned 150K turns cost $0.4700, dropping to $0.1055 in Case A and $0.1419 in Case B. Across every 10-turn 150K coding loop (including output tokens), Thinking Machines: Inkling 2 Small delivers an additional $0.6045 Case-A cost advantage.
Production Workload TCO: 25,000 MAU • 50M Tokens/Mo (50% Cache Hit) • 2,000 GB Egress
💡 Optimal FinOps Verdict: Deploying Thinking Machines: Inkling 2 Small saves $125.40/month ($1505/year) at 25,000 MAU & 50M monthly tokens (80% input / 20% output split with 50% prefix cache hit rate).