Phi 4
Sold by 2 sellers, timed at 100 tok/s, with results on 1 openly licensed board, and not scored by Artificial Analysis.
retrieved sha256 · 0bb53a77713f
The same weights, served by one provider at its own list price, read from deepinfra.com. OpenRouter's rate is its routed rate across whichever sellers it uses, and the two need not agree.
What sellers charge
2 sellers · 1.25x from cheapest to dearestOne tick per seller, on a log scale of input price per million from $0.056 at DeepInfra to $0.07 at DeepInfra. 1 seller undercuts the $0.07 listed; 0 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.
Measured speed
1,127 requests · 2 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
2 priced · 2 timed · cheapest first- DeepInfradeepinfra/bf16/flex$0.056$0.112not listedbf1615knot matchednonenonenot matched100%
- DeepInfradeepinfra/bf16 · listed$0.07$0.14not listedbf1615knot matchednonenonenot matched100%
- DeepInfranot matchednot matchednot matchednot matchedno offer100 tok/s84189ms1,126 reqno offer
- DeepInfranot matchednot matchednot matchednot matchedno offer91 tok/s75827ms1 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 2 timed routes are slower than it, and the middle route runs at 95.5 tok/s. The widest median-to-p90 gap is DeepInfra's, 1.21x.
DeepInfra publishes 2 offers and had 2 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 . 1 of the 2 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: DeepInfra (deepinfra/bf16/flex) at 94.96% on 26 Oct 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
13 parameters · 2 sellers- frequency_penaltyevery seller
- logit_biasevery seller
- max_tokensevery seller
- min_pevery seller
- presence_penaltyevery seller
- repetition_penaltyevery seller
- response_formatevery seller
- seedevery seller
- stopevery seller
- structured_outputsevery seller
- temperatureevery seller
- top_kevery seller
- top_pevery seller
All 13 are accepted by every one of the 2 sellers that publish a list. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.
Price position
342 priced modelsBlended $0.088 per million tokens, 3 in to 1 out: cheaper than 318 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 boardNot ranked in any Design Arena category.
Position 161 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
1 of 6 boards · 3 results| Task | All runs | Score |
|---|---|---|
| GPQA diamond | 342 | 0.56 |
| MATH level 5 | 108 | 0.65 |
| OTIS Mock AIME 2024-2025 | 319 | 0.14 |
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 phi-4, with no effort setting written on it. 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.
1 of the 6 openly licensed boards Vitna mirrors carries a result for Phi 4, each printed as its publisher published it and none added to another. MTEB, LiveBench, 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: each model's own Hugging Face repository (huggingface.co), as of 20 Nov 2026. 162 repositories read, 4 did not answer.
Licence, parameter count, model class and gating: each model's own Hugging Face repository, the one OpenRouter names for the listing.
A licence is the tag the uploader put on the repository, read from nothing else and verified by nobody.
A parameter count is the total the repository's safetensors files declare; for a mixture of experts that is every expert, not the share one token runs through.
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.
