{"slug": "deepseek-v3-0324-cheapest-deepinfra-0-24-m-input", "title": "DeepSeek V3 0324 - cheapest: DeepInfra $0.24/M input", "summary": "DeepSeek V3 0324, a 671B-parameter Mixture-of-Experts model with 37B active per token and a 128k context window, is available on DeepInfra at $0.24 per million input tokens, the cheapest listed provider. The model achieves a general score of 87 and a value score of 363 points per dollar per million input tokens. DeepInfra's uptime is 52.25%, while SiliconFlow offers 98.89% uptime at $0.25 per million input tokens.", "body_md": "# DeepSeek V3 0324\n\nMoE workstation671B total, 37B active per token. Q4_K_M requires ~360GB RAM/VRAM. Use Q2_K (~180GB) for DGX Spark.\n\n- 671.0B\n- 128k\n- mit\n- Mar 2025\n\n## Scores\n\n## Score per dollar\n\n363 pts per $/M input\n\ngeneral_score (87) divided by cheapest input price\n($0.24/M).\nHigher is better value. [See live pricing](/models/deepseek-v3-0324/pricing).\n\n## Run it locally\n\nPer-quant memory needs and a static \"can you run it?\" reference - no rig entry required\n\n### Can you run it? - reference rigs\n\n| Rig | Q2_K | Q4_K_M |\n|---|---|---|\n| NVIDIA Jetson Orin NX 16GB |\n|\n\n[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)[no -> cloud](#cloud-pricing)Fit tiers use the same will-it-run logic as the rig finder. For comfortable fits, the badge reflects decode speed: fast >=20 t/s, ok 8-20 t/s, slow <8 t/s. t/s is a bandwidth estimate, not a measured benchmark.\n\n**How can a 24GB GPU run a 744B model?** It does not load the model into VRAM. The quantized weights (e.g. ~410GB at Q4) sit in system RAM; the GPU only holds the small shared attention and router tensors and accelerates prompt processing. Because GLM 5.2 is a Mixture-of-Experts model, each token activates only ~40B of its 744B params, so llama.cpp streams just those active experts from system RAM to the GPU each token (the `-cmoe`\n\noffload path).\n\nThat makes **decode speed bound by system-RAM bandwidth, not GPU bandwidth** - single digits on DDR4, which is why these rigs show 3-8 t/s even though they “fit.” A bigger GPU (e.g. 2x 3090) keeps more experts resident on-card and raises tok/s; a smaller GPU still runs it but pays the bandwidth tax. A 744B dense model could not run this way - only MoE’s small-active-params trick makes it possible.\n\nAggressive quants (1-2 bit) trade accuracy for size - roughly 17% accuracy loss at 2-bit vs full precision, and real long-context work often needs Q5 or Q6 even when lower quants “fit.”\n\nFormula estimates here are conservative; real tuned setups can exceed them (one HN user reports ~6 tok/s on a 512GB DDR4 + 2x 3090 rig).\n\n## Download options\n\n## Or run it in the cloud\n\nLive per-provider pricing, throughput and uptime - refreshed about 1 hour ago via OpenRouter. Click a column to sort.\n\n| Provider | Type | Input $/M | Output $/M | Cache $/M | Tok/s | Latency | Uptime | Value |\n|---|---|---|---|---|---|---|---|---|\n|\nSiliconFlow\n|\nAPI | 0.25 | 1.00 | 0.135 | - | - | 98.89% | best uptime |\n|\nNovita\n|\nAPI | 0.27 | 1.12 | 0.135 | - | - | 97.79% | |\n|\nDeepInfra\navoid\n|\nAPI | 0.24 | 0.90 | 0.135 | - | - | 52.25% | cheapest |\n\nDefault order: throughput among 95%+ uptime providers, then latency; subscriptions last. Sort by any column. Subscription rows show $/mo in the Value column - per-token columns are \"-\". Affiliate links are marked sponsored / nofollow. Confirm current pricing on the provider's site before committing.\n\n[Detailed API pricing page + JSON endpoint →](/models/deepseek-v3-0324/pricing)\n\n[See who runs DeepSeek in production →](/adoption/deepseek)\n\n## Inference cost over time\n\nData accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.", "url": "https://wpnews.pro/news/deepseek-v3-0324-cheapest-deepinfra-0-24-m-input", "canonical_source": "https://tokenstead.ai/models/deepseek-v3-0324", "published_at": "2026-07-21 21:08:41+00:00", "updated_at": "2026-07-21 21:26:11.383003+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "ai-infrastructure", "ai-products"], "entities": ["DeepSeek V3 0324", "DeepInfra", "SiliconFlow", "Novita", "OpenRouter", "NVIDIA Jetson Orin NX 16GB"], "alternates": {"html": "https://wpnews.pro/news/deepseek-v3-0324-cheapest-deepinfra-0-24-m-input", "markdown": "https://wpnews.pro/news/deepseek-v3-0324-cheapest-deepinfra-0-24-m-input.md", "text": "https://wpnews.pro/news/deepseek-v3-0324-cheapest-deepinfra-0-24-m-input.txt", "jsonld": "https://wpnews.pro/news/deepseek-v3-0324-cheapest-deepinfra-0-24-m-input.jsonld"}}