--- title: "DeepSeek: DeepSeek V4 Pro 0423 — price, capability and what beats it · Undominated.ai" canonical: https://undominated.ai/models/deepseek__deepseek-v4-pro/ description: "DeepSeek: DeepSeek V4 Pro 0423: $0.450/M in, $3.48/M out. Independent capability scores, and alternatives compared under recorded workload and capability requirements." --- # DeepSeek: DeepSeek V4 Pro 0423 — price, capability and what beats it · Undominated.ai > DeepSeek: DeepSeek V4 Pro 0423: $0.450/M in, $3.48/M out. Independent capability scores, and alternatives compared under recorded workload and capability requirements. [Check base rates for this workload](/check/deepseek__deepseek-v4-pro/) [Cheaper alternatives](/alternatives/deepseek__deepseek-v4-pro/) [Family](/families/deepseek/) [Compare models](https://undominated.ai/compare/?models=deepseek%2Fdeepseek-v4-pro) Comparison context Choose up to three standard models in Compare. Requirements and prompt length apply in Compare. This page keeps its stated price basis and workload controls. MODEL PROOF # DeepSeek V4 Pro 0423 DeepSeek · 24 Apr 2026 Balanced Summarise Chat Code gen Agentic BETTER VALUE OPTION ## A stored alternative has an equal-or-higher measured score and an equal-or-lower price. **$1.21** / 1M tokens Balanced · 3 tokens in per 1 out Serving precision differs between offers. [GLM 5.3 Flash](/models/z-ai__glm-5.3-flash/) · Δ score 24.4 points · 80% lower measured price · $0.238 / 1M tokens [MiMo-V2.6-Pro](/models/xiaomi__mimo-v2.6-pro/) is a further stored option with these losses: 384K → 131K max output Evidence as of 06 Oct 2026 [openrouter/models](https://openrouter.ai/api/v1/models) [Vendor cross-check](https://api-docs.deepseek.com/quick_start/pricing) [LMArena · cc-by-4.0](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset) Independent LMArena score 1,445.2 ± 6.59 · 9,780 votes Context window 1.05M Maximum output 384K Input modalities text Output modalities text Published input price $0.450 / 1M tokens Published output price $3.48 / 1M tokens Pricing kind fixed Inspect [complete billing conditions](#billing-details) and endpoint terms below. Price from Parasail: the cheapest offer that is serving, at standard delivery and at a declared precision that is not 4-bit (fp8), at its standard rate. The maker’s own precision is not known. Cheaper right now: $0.261/M at StreamLake — promotion 88% It does not pass the like-for-like test, so it never sets a rank. Every price dimension · USD per million tokens | Input | $0.450 /M | | --- | --- | | Output | $3.48 /M | | Cached input | $0.100 /M | | Cache write | Unknown /M | | Cache write 1h | Unknown /M | | Reasoning | Unknown /M | ### Who sells it The same weights, different shops. Cheapest is not like-for-like when serving precision differs. Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.450/M from Parasail. 15 provider offers · rates, limits and conditions Rates are USD per million tokens. Endpoint terms can differ even when the quoted price and precision match. A listed parameter is a provider declaration, not a task-success test. | Seller | Input | Output | Precision | Uptime (1d) | | --- | --- | --- | --- | --- | | Seller Relace relace/fp4 Endpoint terms Cached input /M $0.210 Context limit 1,048,576 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 97.62% | Input $0.207 | Output $4.20 | Precision fp4 | Uptime (1d) 99.85% | | Seller StreamLake streamlake/fp8 Endpoint terms Cached input /M $0.017 Context limit 1,024,000 tokens Output limit 384,000 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount 88% · already included in these rates Uptime · last 30 minutes 99.60% | Input $0.209 | Output $0.418 | Precision fp8 | Uptime (1d) 98.92% | | Seller Parasail parasail/fp8 Endpoint terms Cached input /M $0.100 Context limit 1,048,576 tokens Output limit 943,718 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 96.73% | Input $0.450 | Output $3.48 | Precision fp8 | Uptime (1d) 98.64% | | Seller GMICloud gmicloud/fp8 Endpoint terms Cached input /M $0.080 Context limit 1,048,576 tokens Output limit 943,718 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount 45% · already included in these rates Uptime · last 30 minutes 95.15% | Input $0.957 | Output $1.91 | Precision fp8 | Uptime (1d) 97.03% | | Seller DigitalOcean digitalocean Endpoint terms Cached input /M $0.209 Context limit 1,048,576 tokens Output limit 384,000 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 96.44% | Input $1.04 | Output $2.09 | Precision undeclared | Uptime (1d) 99.36% | | Seller Cloudflare cloudflare Endpoint terms Cached input /M $0.200 Context limit 1,048,576 tokens Output limit 943,718 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 92.01% | Input $1.15 | Output $2.55 | Precision undeclared | Uptime (1d) 98.19% | | Seller Reka reka Endpoint terms Cached input /M $0.240 Context limit 1,048,576 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.85% | Input $1.20 | Output $12.00 | Precision undeclared | Uptime (1d) 97.97% | | Seller DeepInfra deepinfra/fp8 Endpoint terms Cached input /M $0.100 Context limit 1,048,576 tokens Output limit 16,384 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.90% | Input $1.30 | Output $2.60 | Precision fp8 | Uptime (1d) 99.82% | | Seller Alibaba alibaba/fp8 Endpoint terms Cached input /M $0.118 Context limit 1,000,000 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 95.99% | Input $1.42 | Output $2.83 | Precision fp8 | Uptime (1d) 97.85% | | Seller SiliconFlow siliconflow/fp8 Endpoint terms Cached input /M $0.135 Context limit 1,048,576 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.96% | Input $1.50 | Output $3.14 | Precision fp8 | Uptime (1d) 99.56% | | Seller Novita novita/fp8 Endpoint terms Cached input /M $0.135 Context limit 1,048,576 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.91% | Input $1.60 | Output $3.20 | Precision fp8 | Uptime (1d) 99.97% | | Seller Venice venice Endpoint terms Cached input /M $0.330 Context limit 1,000,000 tokens Output limit 32,768 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 100.00% | Input $1.65 | Output $3.30 | Precision undeclared | Uptime (1d) 99.29% | | Seller AtlasCloud atlas-cloud/fp4 Endpoint terms Cached input /M $0.130 Context limit 1,048,576 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.26% | Input $1.68 | Output $3.38 | Precision fp4 | Uptime (1d) 98.99% | | Seller Baidu baidu/fp8 Endpoint terms Cached input /M $0.140 Context limit 1,048,576 tokens Output limit 393,216 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.85% | Input $1.69 | Output $3.38 | Precision fp8 | Uptime (1d) 99.91% | | Seller Azure azure/us Endpoint terms Cached input /M $0.160 Context limit 1,048,576 tokens Output limit 384,000 tokens Tools Listed by endpoint Reasoning Listed by endpoint Promotional discount None reported Uptime · last 30 minutes 99.83% | Input $1.91 | Output $3.83 | Precision undeclared | Uptime (1d) 99.41% | ### Independent scores | LMArena Elo | 1445.2 high | | --- | --- | LMArena Elo under CC BY 4.0. Absence of another board is not a score of zero. #### Best rankings by task - 3D #24 1262 - Godot games #32 1059 - Game development #36 1242 - Code categories #37 1246 - ASCII art #40 1160 - Web apps #41 1000 - Websites #42 1241 - Full-stack apps #43 948 Per-task Elo and rank from [Design Arena](https://designarena.ai), via OpenRouter’s model feed. The rank is the feed’s, and it counts models OpenRouter does not list. ### Capability | Context window | 1M | | --- | --- | | Max output | 384K | | Input modes | text | | Tool use | yes | | Reasoning | optional | | Open weights | yes | ### Provenance | Price source | [openrouter.ai](https://openrouter.ai/api/v1/models) | | --- | --- | | Fetched | 2026-10-06 | | Quality data | verified | | Cross-checked | [vendor page](https://api-docs.deepseek.com/quick_start/pricing) | ## Our take editorial — not a measurement For scheduled or cache-heavy work, confirm the selected seller’s billing rules before estimating savings. A hosted listing does not establish another endpoint’s time-of-day tariff. The accepted record lists open weights, but deployment still needs a licence and hardware review. Use the current price table and proof for supported measurements, and confirm the tariff with the endpoint you intend to use. ### Strengths - Open weights are recorded - Text input, tool use and cached-input pricing are listed ### Weaknesses - Time-of-day terms must be established for the actual endpoint - A missing licence detail needs investigation before self-hosting ### Reach for it when - Cache-heavy text workload evaluations - Self-hosting assessments with a verified licence ### Avoid it if - Your savings calculation relies on an unverified schedule discount - You require multimodal input Sources: [api-docs.deepseek.com](https://api-docs.deepseek.com/quick_start/pricing) · [openrouter.ai](https://openrouter.ai/api/v1/models) DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,... ## Questions this page answers What does DeepSeek V4 Pro 0423 cost? $0.450/M in, $3.48/M out in this catalogue, as of the fetch date on this page. That is the API row, not a subscription. Does DeepSeek V4 Pro 0423 have an independent quality score? DeepSeek V4 Pro 0423 has an LMArena score in this catalogue. What beats DeepSeek V4 Pro 0423? GLM 5.3 Flash has an equal-or-higher measured score and an equal-or-lower price. Does this page use Artificial Analysis scores for DeepSeek V4 Pro 0423? No. Artificial Analysis figures are not published here. Quality on this page is LMArena Elo where a score exists; otherwise the row is unrated. Is the cheapest DeepSeek V4 Pro 0423 endpoint the same product? Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.450/M from Parasail. MODEL MONUMENT ## Save or share this proof Download proof card · 1200 × 630 Preview and more sharing options DeepSeek: DeepSeek V4 Pro 0423. Balanced workload. Better value option available. Evidence as of 06 Oct 2026. Copy model link Download square card · 1080 × 1080 Copy landscape image Share landscape card ## Related - Both better and cheaper [GLM 5.3 Flash](/models/z-ai__glm-5.3-flash/) - Higher score and cheaper, with a named trade [MiMo-V2.6-Pro](/models/xiaomi__mimo-v2.6-pro/) - Nearby on the capability axis [Kimi K2.5](/models/moonshotai__kimi-k2.5/) - Same vendor [DeepSeek V4 Flash 0423](/models/deepseek__deepseek-v4-flash/) - DeepSeek models [DeepSeek](/providers/deepseek/) - Where this model appears [wire](/wire/) - Nearby on the capability axis [Gemma 4 31B](/models/google__gemma-4-31b-it/) - Same vendor [R1 0528](/models/deepseek__deepseek-r1-0528/) - Where this model appears [self-host](/self-host/) - Nearby on the capability axis [GPT-6.1 Sol](/models/openai__gpt-6.1-sol/) - Same vendor [DeepSeek V4.1 Flash](/models/deepseek__deepseek-v4.1-flash/) - Where this model appears [spreads](/spreads/) ## Badge The verdict as one SVG, for a README or a docs page. It states this model's dominance status at the balanced workload on the LMArena lens, with the date it was computed, and it is rebuilt with the catalogue — so it changes when the verdict changes, including to one you would rather it did not. Markdown Copy [![DeepSeek V4 Pro 0423 — Undominated.ai dominance verdict](https://undominated.ai/badge/deepseek__deepseek-v4-pro.svg)](https://undominated.ai/models/deepseek__deepseek-v4-pro/) HTML Copy DeepSeek V4 Pro 0423 — Undominated.ai dominance verdict Direct file: [https://undominated.ai/badge/deepseek__deepseek-v4-pro.svg](https://undominated.ai/badge/deepseek__deepseek-v4-pro.svg) — a static SVG, written by scripts/build-badges.mjs on every build, so the copy you embed is never older than the last deploy. ## Continue your investigation - [Explore model families](/families/) - [Compare a shortlist](/compare/) - [Check a model](/check/) - [Understand the evidence](/methodology/)