GPT-5.5 (batch)
Sold by 4 sellers, timed at 45.2 tok/s, scored 74.9 on the Coding Index, on 21 Design Arena boards, and with results on 2 openly licensed boards.
retrieved sha256 · 0bb53a77713f
What sellers charge
4 sellers · 5x from cheapest to dearestOne tick per seller, on a log scale of input price per million from $2.50 at OpenAI to $12.50 at OpenAI. The $2.50 listed is the cheapest on offer; 3 sellers are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.
Measured speed
1,722 requests · 5 endpoints · 30-minute windowOpenRouter's own timing of its routed traffic, weighted by requests, over one 30-minute window read . Every route is on the endpoint ledger below.
Every endpoint
5 priced · 5 timed · cheapest first- OpenAIopenai/flex · listed$2.50$15.00$0.25not declared128knot matchednonenonenot matched100%
- Azureazure$5.00$30.00$0.50not declared128k60 tok/s29.83.56s293 req99.36%
- Azureazure$5.00$30.00$0.50not declared128k46 tok/s352.32s6 req99.36%
- OpenAIopenai$5.00$30.00$0.50not declared128knot matchednonenonenot matched100%
- OpenAIopenai/fast$12.50$75.00$1.25not declared128knot matchednonenonenot matchedno reading
- OpenAInot matchednot matchednot matchednot matchedno offer62 tok/s39592ms2 reqno offer
- OpenAInot matchednot matchednot matchednot matchedno offer71.5 tok/s36.91.01s4 reqno offer
- OpenAInot matchednot matchednot matchednot matchedno offer42 tok/s273.95s1,417 reqno offer
Each track runs from a route's p90, the rate nine in ten of its requests met or beat, up to its median. Weighted by requests, the figure above reports where the traffic went rather than what one route does: 1 of the 5 timed routes are slower than it, and the middle route runs at 60 tok/s. The widest median-to-p90 gap is Azure's, 2.01x.
OpenAI publishes 3 offers and had 3 routes timed. The two documents share no route id, so those offers and routes are printed apart rather than paired.
Uptime is OpenRouter's own figure for each seller over the trailing day, read once a day, and the column prints the latest reading, taken . 2 of the 7 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: Azure (azure) at 96.71% on 17 Nov 2026 and OpenAI (openai) at 97.77% on 10 Nov 2026. Source: OpenRouter's public endpoints API, one reading a day for each seller: 1,211 sellers carried one on the latest of the 23 days held.
Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. List prices, not what anybody paid; a discount is the seller's own. Read . Speed is OpenRouter's own timing over one 30-minute window, read .
What each seller accepts
11 parameters · 4 sellers- include_reasoningevery seller
- reasoningevery seller
- reasoning_effortevery seller
- response_formatevery seller
- seedevery seller
- structured_outputsevery seller
- tool_choiceevery seller
- toolsevery seller
- verbosityevery seller
- max_tokens3 of 4not Azure
- max_completion_tokens1 of 4not OpenAI (openai), OpenAI (openai/fast), and OpenAI (openai/flex)
9 of these 11 are accepted by every one of the 4 sellers that publish a list; the other 2 are declared by some sellers and not others, so what a request sending one of them gets depends on which seller serves it. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.
Price position
342 priced modelsBlended $5.625 per million tokens, 3 in to 1 out: cheaper than 35 of the 342 priced models in the catalog. Each bar is one stretch of the catalog on a logarithmic price scale; the mark is this listing.
Measured quality
145 models with an index · 132 on a design arena boardEach mark is one board this model stands on, placed at the share of that board it scores above, and led to its number in the list under the rule. The columns and the ledger below carry the figures.
- Agentic game development · Agents1124 Elowin rate 52.2% · source rank 15the board, 25 ranked
- Agentic HTML slides · Agents1084 Elowin rate 34.2% · source rank 8the board, 9 ranked
- Agentic slides · Agents1150 Elowin rate 43.5% · source rank 6the board, 9 ranked
- Agentic slides, HTML · Agents1077 Elowin rate 33.2% · source rank 8the board, 9 ranked
- Agentic slides, python-pptx · Agents1155 Elowin rate 45.2% · source rank 6the board, 9 ranked
- Android native · Agents1153 Elowin rate 50.8% · source rank 27the board, 41 ranked
- Full stack · Agents1063 Elowin rate 43.0% · source rank 34the board, 44 ranked
- Godot game development · Agents1142 Elowin rate 52.3% · source rank 22the board, 34 ranked
- HTML slides · Agents964 Elowin rate 35.6% · source rank 24the board, 25 ranked
- Mobile apps · Agents1134 Elowin rate 50.4% · source rank 36the board, 49 ranked
- PPTX slides · Agents1157 Elowin rate 45.3% · source rank 6the board, 8 ranked
- python-pptx slides · Agents1152 Elowin rate 43.3% · source rank 21the board, 29 ranked
- Web apps · Agents1097 Elowin rate 42.6% · source rank 39the board, 47 ranked
- 3D scenes · Models1175 Elowin rate 48.8% · source rank 60the board, 113 ranked
- ASCII art · Models1235 Elowin rate 57.2% · source rank 21the board, 67 ranked
- Code · Models1248 Elowin rate 52.8% · source rank 44the board, 120 ranked
- Data visualisation · Models1258 Elowin rate 55.4% · source rank 31the board, 110 ranked
- Game development · Models1281 Elowin rate 56.1% · source rank 23the board, 120 ranked
- SVG · Models1225 Elowin rate 54.7% · source rank 22the board, 83 ranked
- UI components · Models1248 Elowin rate 53.2% · source rank 44the board, 116 ranked
- Websites · Models1247 Elowin rate 52.2% · source rank 45the board, 126 ranked
Position 22 of the 258 models Epoch has placed, which is not the population of either board above, on a scale anchored at Claude 3.5 Sonnet 130 and GPT-5 150. Epoch states that absolute values are meaningless by themselves and that comparisons between models are what the index is for, that values are not linearly related to benchmark accuracy, and that a highly specialised model may score low while being capable in its own domain.
Benchmark results
2 of 6 boards · 44 results| Task | All runs | Effort | Score |
|---|---|---|---|
| Chess Puzzles | 167 | xhigh | 0.63 |
| low | 0.30 | ||
| none | 0.09 | ||
| EBR-bench | 34 | xhigh | 0.34 |
| FrontierMath-Erdos | 16 | xhigh | 0 |
| FrontierMath-Tier-4-v2-Private | 84 | xhigh | 0.72 |
| FrontierMath-Tiers-1-3-v2-Private | 128 | xhigh | 0.85 |
| Furniture Assembly | 42 | xhigh | 0.44 |
| GPQA diamond | 342 | low | 0.91 |
| none | 0.77 | ||
| MirrorCode | 48 | xhigh | 0.23 |
| high | 0.10 | ||
| Mystery Game Puzzles | 149 | xhigh | 0.56 |
| high | 0.52 | ||
| low | 0.28 | ||
| none | 0.18 | ||
| OEIS Open | 3 | medium | 0.26 |
| OEIS Open Lite | 5 | medium | 0.36 |
| OTIS Mock AIME 2024-2025 | 319 | low | 0.84 |
| none | 0.58 | ||
| SimpleQA Verified | 106 | xhigh | 0.63 |
Mean accuracy, 0 to 1, on the task the row names, as Epoch AI runs it and averaged over repeated runs. All eighteen of their tasks, including the held-out FrontierMath tiers, named as they name them. The model id carries the effort setting the run used. Epoch AI Benchmarking Hub names these runs gpt-5.5_xhigh, gpt-5.5_high, gpt-5.5_medium, gpt-5.5_low and gpt-5.5_none, and the effort beside each score says which run it is. A run missing from a task's rows has no published result there, which is not a score of zero. All runs is how many runs the publisher scored on the task, every model's. Read from epoch.ai ↗, which dates its own data . Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
| Task | Score |
|---|---|
| AMPS_Hard | 97 |
| code_completion | 82.61 |
| code_generation | 81.69 |
| connections | 100 |
| consecutive_events | 88.82 |
| integrals_with_game | 99 |
| javascript | 63.64 |
| logic_with_navigation | 76 |
| math_comp | 96.08 |
| olympiad | 91.35 |
| paraphrase | 71.73 |
| plot_unscrambling | 74.09 |
| python | 55 |
| simplify | 66.32 |
| spatial | 98 |
| story_generation | 72.55 |
| summarize | 72.32 |
| tablejoin | 55.90 |
| tablereformat | 100 |
| theory_of_mind | 84.61 |
| typescript | 43.33 |
| typos | 88 |
| zebra_puzzle | 100 |
Percent correct on the task the row names, higher is better. All 23 published columns; there is no average to carry, because the site computes its headline in the browser from a weighting this does not reimplement. LiveBench names this run gpt-5.5-xhigh. Read from livebench.ai ↗, which dates its own data . LiveBench, Apache-2.0. Results produced by the LiveBench maintainers on questions refreshed monthly.
2 of the 6 openly licensed boards Vitna mirrors carry a result for GPT-5.5 (batch), each printed as its publisher published it and none added to another. MTEB, HELM, Vectara hallucination leaderboard and Aider polyglot have not published one under a name that matches its listing, and names are matched to the catalog exactly and never guessed. A task that is a link opens its column on /capability, with every run scored on it; an effort that is a link opens its run's row there, where the run is ranked on Vitna's scale. Every board, with its licence and what its publisher measures, is on /benchmarks.
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.
Source: the Epoch AI Capabilities Index (epoch.ai/eci), as of 20 Nov 2026. 258 models placed over 2917 benchmark results.
Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
Vitna ran none of these evaluations and adjusted none of these figures.
