Last refreshed: November 5, 2026
Update history: August 19, 2026 — first published. October 20, 2026 — Claude figures refreshed (checked October 8, 2026); OpenAI rows carried as October 1 figures, flagged on the page as not re-verified. November 5, 2026 — full rebuild: both vendors verified against current pricing, gpt-6-astra added as OpenAI’s flagship, GPT-5 marked deprecated, legacy context figures corrected. (Later the same day: the two context corrections were themselves reversed — a primary-source read showed Opus 5 and Sonnet 5 carry 1M context; the figures stand as originally published. See the wave-25 log.)
Claude wins on coding quality and long-context reliability. OpenAI’s deprecated GPT-5 was the cheaper mid-tier list; its current mid-tier matches Sonnet 5.5 on price. Pick the SKU, not the family name.
Claude figures verified November 5, 2026 against platform.claude.com pricing and Haiku 5.5. OpenAI figures verified November 5, 2026 against current OpenAI pricing; GPT-5 is deprecated and shuts down December 11, 2026.
The Models Being Compared
Compare Claude Haiku 5.5 / Sonnet 5.5 / Opus 5.5 against gpt-6-astra / gpt-6.1-sol. GPT-5 remains listed below, deprecated. Fable 5.1 sits above Opus on Anthropic’s list price.
| Model | Provider | Input (per 1M tokens) | Output (per 1M tokens) | Context |
|---|---|---|---|---|
| Claude Haiku 5.5 (≤100k / >100k) | Anthropic | $0.10 / $0.50 | $0.50 / $2.50 | priced by prompt length |
| Claude Haiku 4.5 | Anthropic | $1 | $5 | still listed |
| Claude Sonnet 5.5 | Anthropic | $2 | $10 | 1M tokens; cache read $0.10 |
| Claude Sonnet 5 | Anthropic | $2 | $10 | 1M tokens; cache read $0.20 |
| Claude Sonnet 4.6 | Anthropic | $3 | $15 | 1M tokens |
| Claude Opus 5.5 | Anthropic | $4 | $20 | 1M tokens |
| Claude Opus 5 / 4.8 | Anthropic | $5 | $25 | 1M tokens |
| Claude Fable 5.1 | Anthropic | $10 | $50 | 1M tokens |
| GPT-5 | OpenAI | $1.25 | $10 | 400K tokens — deprecated; shuts down December 11, 2026 |
| gpt-6.1-sol | OpenAI | $2 | $10 | prompts over 272K tokens bill at $4 / $15 for the whole request; cache read $0.10 |
| gpt-6-astra (current flagship) | OpenAI | $10 | $50 | cache read $1.00 |
Sonnet 5.5 remains $2 / $10, with cache reads at $0.10. Opus 5.5 remains $4 / $20 with cache reads at $0.20. Haiku 5.5 is the new cheap band: $0.10 / $0.50 under 100,000 prompt tokens, $0.50 / $2.50 over. Haiku 4.5 stays $1 / $5. GPT-5’s list ($1.25 / $10) is cheaper per input token than Sonnet 5.5, but GPT-5 is deprecated and shuts down on December 11, 2026. Do not average Haiku 5.5’s two bands.
Coding Performance
Use Claude Opus 5.5 or Sonnet 5.5 for interactive coding and agentic runs. Use Haiku 5.5 when the task is a subagent, summary, or classification and the prompt stays under 100k tokens.
Latency and Context
Claude Sonnet and Opus SKUs publish a 1M token context window at one rate. Haiku 5.5 raises the rate above 100,000 prompt tokens. Opus 5.5 fast mode is $8 / $40 on the Claude API only.
Cost Comparison for Real Workloads
On Claude, cache reads are now 5% of input on Sonnet 5.5 ($0.10) and Opus 5.5 ($0.20), and 2.5% on Fable 5.1 ($0.25). Haiku 5.5 cache reads are $0.01 under 100k. The 50% batch API still applies. For stateless high-frequency calls, Haiku 5.5 under 100k is the Claude list to beat. gpt-6.1-sol short-context matches Sonnet 5.5 at $2 / $10 (verified November 5, 2026); over the 272K band it bills $4 / $15 for the whole request.
Which API to Choose
Use Claude for coding, long-context document work, and agentic workflows. Use Haiku 5.5 for the cheap loop. Running both vendors is normal.
Related: Claude API token rates.
Get The Signal
What we actually ran, what it measured, and what it means for a restoration shop — every week, including the flops.
Leave a Reply