Get started

DeepSeek V4 Flash 0423

Sold by 13 sellers, timed at 81 tok/s, scored 56.2 on the Coding Index, on 8 Design Arena boards, and with results on 1 openly licensed board.

retrieved sha256 · 0bb53a77713f

Models/DeepSeek/DeepSeek V4 Flash 0423USD per 1M tokens · as OpenRouter routes it
deepseek/deepseek-v4-flashDeepSeek · base listing · DeepSeek tokenizer
OpenRouter list price · the routed rate, as the catalog carried itretrieved 17 Nov 16:29 UTC from openrouter.ai ↗
Input$0.09per 1M tokens
Output$0.18per 1M tokens
Cached input$0.018per 1M tokens
Context1.05M1,048,576 tokens
What Vitna read · 2 providers, each from its own pricing page, kept as dated recordsin / out per 1M tokens
DeepInfraopen weights$0.09 / $0.18confirmed 22 Nov 2026
Parasailopen weights$0.14 / $0.28confirmed 22 Nov 2026

The same weights served by 2 providers, each at its own list price. OpenRouter's rate is its routed rate across whichever of them it uses, and it need not match any one of them.

What the list says · the fields OpenRouter publishes for this id10 fields
AuthorDeepSeek
Listed
Canonical iddeepseek/deepseek-v4-flash-20260423
Context1,048,576 tokens
Max output393,216 tokens
Modalitiestext → text
TokenizerDeepSeek
Reasoningoptional · effort xhigh, high, none, high by default
Moderationnot moderated by OpenRouter
What the weights say · read from the model's own repository on Hugging Faceretrieved from huggingface.co ↗
Licencemit
Parameters291B · 290,944,616,402 declared
Model classDeepseekV4ForCausalLM
Downloadnot gated
Published

What sellers charge

13 sellers · 15.22x from cheapest to dearest

One tick per seller, on a log scale of input price per million from $0.0138 at Relace to $0.21 at Azure. 5 sellers undercut the $0.09 listed; 7 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.

The same sellers grouped by the number format each declares, on the same scale. The fp8 group alone spans 92% of the range, so the cheap end is not cheap for its number format. A declaration is checked by nobody: this groups the rows and certifies none.

Measured speed

708,026 requests · 13 endpoints · 30-minute window
Throughput81 tok/sMedian output rate, weighted across the 13 endpoints serving it by the requests each answered.the speed board
First token2.04sMedian wait before the first token, over the same window and weighted the same way.how Vitna reads a figure
Endpoint spread21 to 125The slowest and the fastest route serving this model, in tokens per second. Speed is a property of the route, not of the weights.the board, fastest first

OpenRouter'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

13 priced · 13 timed · cheapest first
  1. Relacerelace/fp4$0.0138$0.45$0.0138fp4944k120 tok/s62.1462ms71,457 req99.44%
  2. Baidubaidu/fp8 · 90% off$0.01456$0.02912$0.002912fp8131k83 tok/s52822ms174,400 req100%
  3. StreamLakestreamlake/fp8 · 80% off$0.028$0.056$0.0056fp8384k21 tok/s137.24s153,695 req98.81%
  4. Fireworksfireworks$0.07$0.14$0.014not declared944k80 tok/s56522ms12,505 req90.15%
  5. DeepInfradeepinfra/fp8/flex$0.072$0.144$0.0144fp8393knot matchednonenonenot matched100%
  6. DeepInfradeepinfra/fp8 · listed$0.09$0.18$0.018fp8393knot matchednonenonenot matched100%
  7. Waferwafer/mxfp4$0.12$0.17$0.024mxfp4944k84 tok/s50341ms65,711 req99.86%
  8. SiliconFlowsiliconflow/fp8$0.13$0.28$0.028fp8393k59 tok/s331.02s2,296 req97.98%
  9. DeepInfradeepinfra/fp8/priority$0.135$0.27$0.027fp8393knot matchednonenonenot matched100%
  10. Venicevenice$0.138$0.275$0.028not declared262k105 tok/s51735ms23,575 req100%
  11. Parasailparasail/fp8$0.14$0.28$0.07fp8944k56 tok/s18719ms13,608 req100%
  12. Mancer 2mancer/fp8$0.18$0.50$0.07fp8944k91 tok/s46641ms17,816 req99.80%
  13. Azureazure/us$0.21$0.56$0.031not declared384k120 tok/s581.03s2,438 req100%
  14. DeepInfranot matchednot matchednot matchednot matchedno offer70.5 tok/s28.91.92s117 reqno offer
  15. DeepInfranot matchednot matchednot matchednot matchedno offer125 tok/s66448ms51,350 reqno offer
  16. DeepInfranot matchednot matchednot matchednot matchedno offer108 tok/s56520ms119,058 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: 5 of the 13 timed routes are slower than it, and the middle route runs at 84 tok/s. The widest median-to-p90 gap is Parasail's, 3.11x.

DeepInfra 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 . 12 of the 19 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: Alibaba at 0.00% on 2 Nov 2026, Fireworks at 52.42% on 9 Nov 2026, Baidu at 69.02% on 10 Nov 2026, GMICloud at 81.95% on 10 Nov 2026, and 8 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

22 parameters · 13 sellers
  1. include_reasoningevery seller
  2. max_tokensevery seller
  3. reasoningevery seller
  4. reasoning_effortevery seller
  5. response_formatevery seller
  6. temperatureevery seller
  7. tool_choiceevery seller
  8. toolsevery seller
  9. top_pevery seller
  10. stop11 of 13not Azure and SiliconFlow
  11. frequency_penalty10 of 13not Azure, Baidu, and StreamLake
  12. structured_outputs10 of 13not Azure, Relace, and SiliconFlow
  13. top_k10 of 13not Azure, Baidu, and StreamLake
  14. presence_penalty9 of 13
  15. logit_bias8 of 13
  16. repetition_penalty8 of 13
  17. min_p6 of 13
  18. seed6 of 13
  19. logprobs5 of 13
  20. top_logprobs5 of 13
  21. max_completion_tokens1 of 13only Azure
  22. top_a1 of 13only Mancer 2

9 of these 22 are accepted by every one of the 13 sellers that publish a list; the other 13 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 models

Blended $0.113 per million tokens, 3 in to 1 out: cheaper than 307 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 board

Each 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.

Intelligence Indexnot scoredArtificial Analysis has not scored this model on its Intelligence Index.the intelligence leaderboard
Coding Index56.2Position 43 of 118, above 64% of the models scored.the coding leaderboard
Agentic Index22.2Position 40 of 94, above 57% of the models scored.the agentic leaderboard
  1. 3D scenes · Models1169 Elowin rate 49.3% · source rank 64the board, 113 ranked
  2. ASCII art · Models1093 Elowin rate 42.7% · source rank 63the board, 67 ranked
  3. Code · Models1194 Elowin rate 48.9% · source rank 66the board, 120 ranked
  4. Data visualisation · Models1124 Elowin rate 40.7% · source rank 93the board, 110 ranked
  5. Game development · Models1186 Elowin rate 50.2% · source rank 64the board, 120 ranked
  6. SVG · Models1144 Elowin rate 48.4% · source rank 52the board, 83 ranked
  7. UI components · Models1153 Elowin rate 44.7% · source rank 82the board, 116 ranked
  8. Websites · Models1198 Elowin rate 49.1% · source rank 68the board, 126 ranked
Epoch Capabilities Index · a fit by a third party over every benchmark this model has results onthe whole index, with intervals
Index146.9143.7 to 150.2
Position87 of 258above 66% of the models placed
Results behind it4benchmark results
As Epoch names itDeepSeek-V4-Flashreleased 24 Apr 2026

Position 87 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 · 23 results
LiveBench · publisher-run, Apache-2.023 of 23 tasks · 71 runs scored on each task · effort unstated

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 deepseek-v4-flash, with no effort setting written on it. Read from livebench.ai ↗, which dates its own data . LiveBench, Apache-2.0. Results produced by the LiveBench maintainers on questions refreshed monthly.

1 of the 6 openly licensed boards Vitna mirrors carries a result for DeepSeek V4 Flash 0423, each printed as its publisher published it and none added to another. Epoch AI Benchmarking Hub, 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.

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.
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.