Get started

Claude Opus 5 against Claude Opus 5.5.

Two listings from the same committed catalog, side by side: the fields OpenRouter publishes, the boards two third parties scored, and a cost sheet that is arithmetic over the listed rates.

retrieved sha256 · 0bb53a77713f

Compare/Claude Opus 5vsClaude Opus 5.5USD per 1M tokens · one scale
Listed
1,000,000 tokens
Context
1,000,000 tokens
128,000 tokens
Max output
128,000 tokens
text, image, file → text
Modalities
text, image, file → text
not stated
Knowledge cutoff
not stated
57.3 tok/s
Throughput
80.8 tok/s
1.21s
First token
2.1s
6,197 requests · 9 endpoints
Timed over
134,352 requests · 11 endpoints

Listed price

USD per million tokens · one linear scale
Claude Opus 5Anthropic · listed 24 Jul 2026
Input per M$5.00
Cached input per M$0.50
Output per M$25.00
Blended, 3 in to 1 out$10.00
Claude Opus 5.5Anthropic · listed 22 Sep 2026
Input per M$4.00
Cached input per M$0.20
Output per M$20.00
Blended, 3 in to 1 out$8.00

Per million tokens, as OpenRouter listed them when the catalog was read: its routed rates, not a provider's own page. One linear scale across every row, so a bar twice as long is twice the price. The blended row is Vitna's arithmetic, three input tokens to one output.

Every board either stands on

22 boards
44.2 · 12 of 58
Intelligence Index
59.1 · 03 of 58
78.0 · 03 of 118
Coding Index
not scored
56.5 · 02 of 94
Agentic Index
not scored
1303 · 16 of 126
Websites · Models
1325 · 06 of 126
1316 · 13 of 120
Code · Models
1384 · 02 of 120
1329 · 13 of 120
Game development · Models
1427 · 01 of 120
1339 · 11 of 116
UI components · Models
1375 · 04 of 116
1322 · 15 of 113
3D scenes · Models
1462 · 02 of 113
1326 · 09 of 110
Data visualisation · Models
1385 · 01 of 110
1305 · 07 of 83
SVG · Models
1321 · 05 of 83
1338 · 04 of 67
ASCII art · Models
1349 · 03 of 67
1348 · 01 of 49
Mobile apps · Agents
1294 · 03 of 49
1248 · 12 of 47
Web apps · Agents
1319 · 02 of 47
1280 · 07 of 44
Full stack · Agents
1391 · 02 of 44
1245 · 08 of 41
Android native · Agents
1389 · 01 of 41
1249 · 09 of 36
pdf · Agents
1316 · 02 of 36
1250 · 03 of 34
Godot game development · Agents
not ranked
1242 · 11 of 32
spreadsheet · Agents
1291 · 06 of 32
1224 · 07 of 29
python-pptx slides · Agents
not ranked
1259 · 07 of 26
codetovideo · Agents
1351 · 02 of 26
1202 · 03 of 25
Agentic game development · Agents
not ranked
1201 · 05 of 25
blender3d · Agents
1225 · 04 of 25

Score and position on each board, from the board's own source as OpenRouter carries it. The heavier figure leads that board; a missing score is a fact about coverage, not a zero.

Vitna reads

from provider pages

Where the registry tracks a model, Vitna reads the rate from the provider page itself and keeps it as a dated record; the OpenRouter figures above are its routed rates.

DeepInfra · Claude Opus 5$5.00 / $25.00Anthropic · Claude Opus 5$5.00 / $25.00Anthropic · Claude Opus 5.5$4.00 / $20.00

Cost sheet

800 in · 200 out per call

Arithmetic over the listed rates and nothing else: tokens in over one million times the input rate, plus tokens out over one million times the output rate. No cache, no retries and no failure rate assumed; the Decision Lab is where cost per successful unit lives.

ModelPer callPer 1,000 callsPer 100,000 calls
Claude Opus 5$0.009000$9.00$900
Claude Opus 5.5$0.007200$7.20$720

Claude Opus 5.5 costs 20% less per call at these token counts.

Other pairs with a page
For one of your workloads

The market on these two is here. The right one for your work is not.

What this page cannot say is which is right for a particular piece of your production work, because that depends on what the workload requires and what your own evaluations measured, neither of which is public. The register weighs a candidate against a workload's own thresholds and shows what the engine would decide before anything is filed.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Vitna ran none of these evaluations and adjusted none of these figures.