GPT-6.1 Sol
Sold by 6 sellers, timed at 43 tok/s, scored 54.6 on the Intelligence Index, on 14 Design Arena boards, and with results on 2 openly licensed boards.
retrieved sha256 · 0bb53a77713f
The maker's own list price, read from developers.openai.com. OpenRouter's rate is its routed rate across whichever sellers it uses, and the two need not agree.
What sellers charge
6 sellers · 12x from cheapest to dearestOne tick per seller, on a log scale of input price per million from $1.00 at OpenAI to $12.00 at OpenAI. 1 seller undercuts the $2.00 listed; 2 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.
Measured speed
82,045 requests · 9 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
9 priced · 9 timed · cheapest first- OpenAIopenai/flex$1.00$5.00$0.05not declared128knot matchednonenonenot matched100%
- Amazon Bedrockamazon-bedrock/global · listed$2.00$10.00$0.10not declared128k48 tok/s351.84s4,597 req100%
- Amazon Bedrockamazon-bedrock/global · listed$2.00$10.00$0.10not declared128k47.5 tok/s28.9749ms4 req100%
- Azureazure · listed$2.00$10.00$0.10not declared128k36 tok/s202.43s17,711 req100%
- Azureazure · listed$2.00$10.00$0.10not declared128k37 tok/s264.16s166 req100%
- Azureazure · listed$2.00$10.00$0.10not declared128k41 tok/s301.23s753 req100%
- OpenAIopenai · listed$2.00$10.00$0.10not declared128knot matchednonenonenot matched100%
- OpenAIopenai/fast$4.00$20.00$0.20not declared128knot matchednonenonenot matched100%
- OpenAIopenai/ultrafast$12.00$60.00$0.60not declared128knot matchednonenonenot matched100%
- OpenAInot matchednot matchednot matchednot matchedno offer138 tok/s34404ms1 reqno offer
- OpenAInot matchednot matchednot matchednot matchedno offer49 tok/s244.36s3,613 reqno offer
- OpenAInot matchednot matchednot matchednot matchedno offer69.5 tok/s391.43s1,150 reqno offer
- OpenAInot matchednot matchednot matchednot matchedno offer44 tok/s242.57s54,050 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: 3 of the 9 timed routes are slower than it, and the middle route runs at 47.5 tok/s. The widest median-to-p90 gap is OpenAI's, 4.04x.
OpenAI publishes 4 offers and had 4 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 . 7 of the 10 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 38.68% on 11 Nov 2026, OpenAI (openai/flex) at 39.98% on 23 Oct 2026, Amazon Bedrock (amazon-bedrock/global) at 73.75% on 4 Nov 2026, OpenAI (openai/ultrafast) at 89.88% on 10 Nov 2026, and 3 more. 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 · 6 sellers- include_reasoningevery seller
- reasoningevery seller
- reasoning_effortevery seller
- tool_choiceevery seller
- toolsevery seller
- verbosityevery seller
- max_tokens5 of 6not Azure
- response_format5 of 6not Amazon Bedrock
- seed5 of 6not Amazon Bedrock
- structured_outputs5 of 6not Amazon Bedrock
- max_completion_tokens1 of 6only Azure
6 of these 11 are accepted by every one of the 6 sellers that publish a list; the other 5 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 $4.00 per million tokens, 3 in to 1 out: cheaper than 55 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.
- Android native · Agents1194 Elowin rate 49.0% · source rank 15the board, 41 ranked
- blender3d · Agents1244 Elowin rate 60.8% · source rank 2the board, 25 ranked
- codetovideo · Agents1204 Elowin rate 55.6% · source rank 13the board, 26 ranked
- Mobile apps · Agents1225 Elowin rate 52.9% · source rank 14the board, 49 ranked
- pdf · Agents1131 Elowin rate 43.0% · source rank 29the board, 36 ranked
- spreadsheet · Agents1209 Elowin rate 41.4% · source rank 14the board, 32 ranked
- Web apps · Agents1287 Elowin rate 57.4% · source rank 6the board, 47 ranked
- 3D scenes · Models1426 Elowin rate 59.1% · source rank 5the board, 113 ranked
- Code · Models1361 Elowin rate 56.9% · source rank 4the board, 120 ranked
- Data visualisation · Models1284 Elowin rate 52.7% · source rank 16the board, 110 ranked
- Game development · Models1406 Elowin rate 59.9% · source rank 3the board, 120 ranked
- SVG · Models1336 Elowin rate 59.7% · source rank 3the board, 83 ranked
- UI components · Models1357 Elowin rate 58.8% · source rank 6the board, 116 ranked
- Websites · Models1315 Elowin rate 54.0% · source rank 9the board, 126 ranked
Position 6 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 · 60 results| Task | All runs | Score |
|---|---|---|
| Chess Puzzles | 167 | 0.68 |
| EBR-bench | 34 | 0.54 |
| FrontierMath-Erdos | 16 | 0.04 |
| FrontierMath-Tier-4-v2-Private | 84 | 1 |
| FrontierMath-Tiers-1-3-v2-Private | 128 | 0.94 |
| Furniture Assembly | 42 | 0.80 |
| GPQA diamond | 342 | 0.95 |
| MirrorCode | 48 | 0.61 |
| Mystery Game Puzzles | 149 | 0.80 |
| OTIS Mock AIME 2024-2025 | 319 | 1 |
| SimpleQA Verified | 106 | 0.74 |
| SpireClimb | 12 | 0.24 |
| SpireClimb (Slay the Spire 1, Ascension 20 Heart) | 12 | 0.18 |
| SpireClimb (Slay the Spire 2, Ascension 10) | 12 | 0.29 |
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 this run gpt-6.1-sol_max. 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 | Effort | Score |
|---|---|---|
| AMPS_Hard | max | 98 |
| xhigh | 98 | |
| code_completion | max | 80.44 |
| xhigh | 82.61 | |
| code_generation | max | 80.28 |
| xhigh | 78.87 | |
| connections | max | 100 |
| xhigh | 100 | |
| consecutive_events | max | 90.56 |
| xhigh | 91.29 | |
| integrals_with_game | max | 100 |
| xhigh | 99 | |
| javascript | max | 63.64 |
| xhigh | 63.64 | |
| logic_with_navigation | max | 86 |
| xhigh | 82 | |
| math_comp | max | 97.06 |
| xhigh | 97.06 | |
| olympiad | max | 92.27 |
| xhigh | 91.86 | |
| paraphrase | max | 72.58 |
| xhigh | 71.03 | |
| plot_unscrambling | max | 84.38 |
| xhigh | 81.94 | |
| python | max | 60 |
| xhigh | 60 | |
| simplify | max | 71.97 |
| xhigh | 68.17 | |
| spatial | max | 98 |
| xhigh | 98 | |
| story_generation | max | 70.82 |
| xhigh | 70.57 | |
| summarize | max | 81.25 |
| xhigh | 75.53 | |
| tablejoin | max | 57.42 |
| xhigh | 55.25 | |
| tablereformat | max | 100 |
| xhigh | 100 | |
| theory_of_mind | max | 86.54 |
| xhigh | 86.54 | |
| typescript | max | 40 |
| xhigh | 46.67 | |
| typos | max | 86 |
| xhigh | 84 | |
| zebra_puzzle | max | 100 |
| xhigh | 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 these runs gpt-6.1-sol-max and gpt-6.1-sol-xhigh, and the effort beside each score says which run it is. 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-6.1 Sol, 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.
