# Quels outils IA les entreprises utilisent-elles pour Inférence IA et cloud GPU ?

Baseten: 80 entreprises (26 Confirmé, 20 Déclaré, 34 Observé, 0 Offre d’emploi). Extrait original: “The number that matters to us and our customers is the cost of a finished review. On Baseten, that’s 3.1x lower than with our previous closed model.” — Unblocked ([Lire la source](<https://www.baseten.co/resources/customers/how-unblocked-shifted-to-multi-model-intelligence-for-lower-cost-per-task/>)). · Fireworks AI: 59 entreprises (14 Confirmé, 22 Déclaré, 23 Observé, 0 Offre d’emploi). Extrait original: “leading to a 50% cost reduction and superior answer quality” — Genspark ([Lire la source](<https://fireworks.ai/customers/genspark-deep-research-agent-rft>)). · Nebius AI Cloud: 53 entreprises (30 Confirmé, 0 Déclaré, 23 Observé, 0 Offre d’emploi). Extrait original: “With Nebius’ AI-native infrastructure, they’ve trained a full-transcriptome foundation model (Converge-SC) capable of processing 20,000+ genes per cell — delivering state-of-the-art accuracy, explainability and speed for drug discovery and clinical development.” — Converge ([Lire la source](<https://nebius.com/customer-stories/converge-bio>)).

Périmètre de la réponse: Inférence IA et cloud GPU; ensemble du marché. Voir la réponse citée: https://www.usedby.ai/fr/for?category=llm-inference

# Ce que vos pairs utilisent

80 entreprises sur usedby utilisent Baseten avec une preuve publique (ensemble du marché). T4; 26 CONFIRMED, 20 DECLARED, 34 OBSERVED, 0 SIGNAL. https://www.usedby.ai/fr/for?category=llm-inference

Les comptes portent sur les entreprises recensées par usedby. Ce ne sont pas des parts de marché. Les quatre niveaux de preuve comptent. Une taille inconnue est exclue uniquement des vues par taille. Vos exigences n’ont pas été évaluées.

```json
{
  "status": "answered",
  "category": {
    "slug": "llm-inference",
    "name": "Inférence IA et cloud GPU"
  },
  "industry": null,
  "industryGroup": null,
  "size": null,
  "country": null,
  "answeringTier": "T4",
  "tiersTried": [
    {
      "tier": "T1",
      "label": "même taille",
      "companies": 0,
      "tools": 0,
      "smallSample": true,
      "skipped": true,
      "topTools": []
    },
    {
      "tier": "T2",
      "label": "même secteur",
      "companies": 0,
      "tools": 0,
      "smallSample": true,
      "skipped": true,
      "topTools": []
    },
    {
      "tier": "T3",
      "label": "même groupe de secteurs",
      "companies": 0,
      "tools": 0,
      "smallSample": true,
      "skipped": true,
      "topTools": []
    },
    {
      "tier": "T4",
      "label": "ensemble du marché",
      "companies": 347,
      "tools": 12,
      "smallSample": false,
      "skipped": false,
      "topTools": [
        "Baseten",
        "Fireworks AI",
        "Nebius AI Cloud"
      ]
    }
  ],
  "sample": {
    "companies": 347,
    "strongProofShare": 0.5100864553314121,
    "sizeKnownShare": 0.7896253602305475,
    "excludedVendorCompanies": 3
  },
  "tools": {
    "total": 12,
    "listed": [
      {
        "selectionBasis": "category",
        "does": "Plateforme d'inférence pour déployer et faire évoluer automatiquement des modèles IA open source ou sur mesure.",
        "name": "Baseten",
        "slug": "baseten",
        "companies": 80,
        "mix": {
          "confirmed": 26,
          "declared": 20,
          "observed": 34,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T13:41:28.645Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 70,
          "listed": [
            {
              "usageId": "cmuvwgyoo00apsp1ogc56d7qm",
              "company": "Wayfair",
              "slug": "wayfair",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-20T18:10:03.613Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwgyoo00apsp1ogc56d7qm",
                "method": "LOGO_WALL",
                "original": "https://www.baseten.co/resources/customers/"
              }
            },
            {
              "usageId": "cmvmu0rx80049tc37e60vdnci",
              "company": "Nvidia",
              "slug": "nvidia",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-10-31T06:22:01.739Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmvmu0rx80049tc37e60vdnci",
                "method": "LOGO_WALL",
                "original": "https://www.baseten.co/pricing/"
              }
            },
            {
              "usageId": "cmv2x8hzu00hfsk4c4a3mvtdq",
              "company": "Datadog",
              "slug": "datadog",
              "proof": "DECLARED",
              "size": "ENTERPRISE",
              "lastCheckedAt": "2026-11-22T13:23:23.902Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmv2x8hzu00hfsk4c4a3mvtdq",
                "method": "SUBPROCESSOR_LIST",
                "original": "https://www.datadoghq.com/legal/subprocessors/"
              }
            }
          ],
          "overflow": 67
        },
        "strongProofPeers": {
          "total": 46,
          "listed": [
            {
              "usageId": "cmwpap5xs007jvk1qtybe94kb",
              "company": "Unblocked",
              "slug": "unblocked",
              "proof": "CONFIRMED",
              "size": "STARTUP",
              "lastCheckedAt": "2026-11-20T18:11:33.989Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmwpap5xs007jvk1qtybe94kb",
                "method": "CASE_STUDY",
                "original": "https://www.baseten.co/resources/customers/how-unblocked-shifted-to-multi-model-intelligence-for-lower-cost-per-task/"
              }
            },
            {
              "usageId": "cmuvwfhe8006zsp1o5t4wtbz1",
              "company": "Wispr Flow",
              "slug": "wispr-flow",
              "proof": "CONFIRMED",
              "size": "SMALL",
              "lastCheckedAt": "2026-11-20T18:10:03.613Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwfhe8006zsp1o5t4wtbz1",
                "method": "CASE_STUDY",
                "original": "https://www.baseten.co/resources/customers/wispr-flow/"
              }
            },
            {
              "usageId": "cmuvwfit50073sp1ox84rl09o",
              "company": "OpenEvidence",
              "slug": "openevidence",
              "proof": "CONFIRMED",
              "size": null,
              "lastCheckedAt": "2026-11-20T18:10:03.613Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwfit50073sp1ox84rl09o",
                "method": "CASE_STUDY",
                "original": "https://www.baseten.co/resources/customers/openevidence-baseten-training/"
              }
            }
          ],
          "overflow": 43
        },
        "why": [
          {
            "text": "The number that matters to us and our customers is the cost of a finished review. On Baseten, that’s 3.1x lower than with our previous closed model.",
            "sourceUrl": "https://www.baseten.co/resources/customers/how-unblocked-shifted-to-multi-model-intelligence-for-lower-cost-per-task/",
            "sourceLang": "en",
            "field": "highlightText",
            "usageId": "cmwpap5xs007jvk1qtybe94kb",
            "company": {
              "name": "Unblocked",
              "slug": "unblocked"
            }
          },
          {
            "text": "With Baseten, Flow’s entire pipeline, from speech recognition models to Llama-based transcript enhancement, runs end-to-end in under 700 milliseconds. This latency target isn’t for p50 or even p90 – this is the p99 latency to ensure a smooth user experience.",
            "sourceUrl": "https://www.baseten.co/resources/customers/wispr-flow/",
            "sourceLang": "en",
            "field": "highlightText",
            "usageId": "cmuvwfhe8006zsp1o5t4wtbz1",
            "company": {
              "name": "Wispr Flow",
              "slug": "wispr-flow"
            }
          }
        ],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 20,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published"
      },
      {
        "selectionBasis": "category",
        "does": "Cloud d'inférence et d'entraînement pour servir, affiner et faire passer à l'échelle des modèles open source.",
        "name": "Fireworks AI",
        "slug": "fireworks-ai",
        "companies": 59,
        "mix": {
          "confirmed": 14,
          "declared": 22,
          "observed": 23,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T13:46:33.514Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 48,
          "listed": [
            {
              "usageId": "cmuvwbrlv003bsp1o40r4pfaq",
              "company": "Uber",
              "slug": "uber",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-17T06:41:38.907Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwbrlv003bsp1o40r4pfaq",
                "method": "LOGO_WALL",
                "original": "https://fireworks.ai/"
              }
            },
            {
              "usageId": "cmuvwbrxk003esp1o3plrtbwx",
              "company": "Doordash",
              "slug": "doordash",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-17T06:41:38.907Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwbrxk003esp1o3plrtbwx",
                "method": "LOGO_WALL",
                "original": "https://fireworks.ai/"
              }
            },
            {
              "usageId": "cmuvwbwij0040sp1oxm9o7dci",
              "company": "Samsung",
              "slug": "samsung",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-17T06:41:38.907Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwbwij0040sp1oxm9o7dci",
                "method": "LOGO_WALL",
                "original": "https://fireworks.ai/"
              }
            }
          ],
          "overflow": 45
        },
        "strongProofPeers": {
          "total": 36,
          "listed": [
            {
              "usageId": "cmuvwbra60038sp1odgy86r1w",
              "company": "Genspark",
              "slug": "genspark",
              "proof": "CONFIRMED",
              "size": "SMALL",
              "lastCheckedAt": "2026-11-17T06:42:11.406Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwbra60038sp1odgy86r1w",
                "method": "CASE_STUDY",
                "original": "https://fireworks.ai/customers/genspark-gen-1-slides-rl-post-training"
              }
            },
            {
              "usageId": "cmuvwbmy4002ssp1ojwlpdv13",
              "company": "Cursor",
              "slug": "cursor",
              "proof": "CONFIRMED",
              "size": "MEDIUM",
              "lastCheckedAt": "2026-11-17T06:41:56.023Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwbmy4002ssp1ojwlpdv13",
                "method": "CASE_STUDY",
                "original": "https://fireworks.ai/customers/cursor-fast-apply-speculative-decoding"
              }
            },
            {
              "usageId": "cmuvwbn9u002vsp1oagrpzxcc",
              "company": "Vercel",
              "slug": "vercel",
              "proof": "CONFIRMED",
              "size": "MEDIUM",
              "lastCheckedAt": "2026-11-17T06:41:56.023Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwbn9u002vsp1oagrpzxcc",
                "method": "CASE_STUDY",
                "original": "https://fireworks.ai/customers/vercel-v0-code-fixing-case-study"
              }
            }
          ],
          "overflow": 33
        },
        "why": [
          {
            "text": "leading to a 50% cost reduction and superior answer quality",
            "sourceUrl": "https://fireworks.ai/customers/genspark-deep-research-agent-rft",
            "sourceLang": "en",
            "field": "metricText",
            "usageId": "cmuvwbra60038sp1odgy86r1w",
            "company": {
              "name": "Genspark",
              "slug": "genspark"
            }
          },
          {
            "text": "Fireworks provides the inference and rollout layer that allows Cursor to scale RL without building a dedicated inference system.",
            "sourceUrl": "https://fireworks.ai/customers/cursor-composer-2-reinforcement-learning",
            "sourceLang": "en",
            "field": "highlightText",
            "usageId": "cmuvwbmy4002ssp1ojwlpdv13",
            "company": {
              "name": "Cursor",
              "slug": "cursor"
            }
          }
        ],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 31,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published"
      },
      {
        "selectionBasis": "category",
        "does": "Cloud IA avec clusters GPU NVIDIA non virtualisés, outils MLOps et inférence managée Token Factory.",
        "name": "Nebius AI Cloud",
        "slug": "nebius",
        "companies": 53,
        "mix": {
          "confirmed": 30,
          "declared": 0,
          "observed": 23,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-20T18:23:53.557Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 38,
          "listed": [
            {
              "usageId": "cmuygosbs013osg4cdp6848bv",
              "company": "Revolut",
              "slug": "revolut",
              "proof": "CONFIRMED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-20T18:19:42.584Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuygosbs013osg4cdp6848bv",
                "method": "CASE_STUDY",
                "original": "https://nebius.com/customer-stories/revolut"
              }
            },
            {
              "usageId": "cmwpb0spe0066vk1qfp547osw",
              "company": "TD SYNNEX",
              "slug": "td-synnex",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-20T18:19:21.724Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmwpb0spe0066vk1qfp547osw",
                "method": "LOGO_WALL",
                "original": "https://nebius.com/"
              }
            },
            {
              "usageId": "cmuygorxe013lsg4cpc5xjahj",
              "company": "Shopify",
              "slug": "shopify",
              "proof": "OBSERVED",
              "size": "ENTERPRISE",
              "lastCheckedAt": "2026-11-20T18:19:21.724Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuygorxe013lsg4cpc5xjahj",
                "method": "LOGO_WALL",
                "original": "https://nebius.com/"
              }
            }
          ],
          "overflow": 35
        },
        "strongProofPeers": {
          "total": 30,
          "listed": [
            {
              "usageId": "cmuygp07j0153sg4co3tuq6e7",
              "company": "Converge",
              "slug": "converge",
              "proof": "CONFIRMED",
              "size": "STARTUP",
              "lastCheckedAt": "2026-11-20T18:23:53.557Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuygp07j0153sg4co3tuq6e7",
                "method": "CASE_STUDY",
                "original": "https://nebius.com/customer-stories/converge-bio"
              }
            },
            {
              "usageId": "cmuygq4vf0180sg4czc6vz3ct",
              "company": "SynthLabs",
              "slug": "synthlabs",
              "proof": "CONFIRMED",
              "size": "STARTUP",
              "lastCheckedAt": "2026-11-20T18:23:36.854Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuygq4vf0180sg4czc6vz3ct",
                "method": "CASE_STUDY",
                "original": "https://nebius.com/customer-stories/synthlabs"
              }
            },
            {
              "usageId": "cmuygp3w7015esg4c4n6rnf3y",
              "company": "Quantori",
              "slug": "quantori",
              "proof": "CONFIRMED",
              "size": "MEDIUM",
              "lastCheckedAt": "2026-11-20T18:23:20.304Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuygp3w7015esg4c4n6rnf3y",
                "method": "CASE_STUDY",
                "original": "https://nebius.com/customer-stories/quantori"
              }
            }
          ],
          "overflow": 27
        },
        "why": [
          {
            "text": "With Nebius’ AI-native infrastructure, they’ve trained a full-transcriptome foundation model (Converge-SC) capable of processing 20,000+ genes per cell — delivering state-of-the-art accuracy, explainability and speed for drug discovery and clinical development.",
            "sourceUrl": "https://nebius.com/customer-stories/converge-bio",
            "sourceLang": "en",
            "field": "highlightText",
            "usageId": "cmuygp07j0153sg4co3tuq6e7",
            "company": {
              "name": "Converge",
              "slug": "converge"
            }
          },
          {
            "text": "SynthLabs found TractoAI’s ability to automatically scale up and schedule offline inference jobs with respect to the map operators particularly impactful.",
            "sourceUrl": "https://nebius.com/customer-stories/synthlabs",
            "sourceLang": "en",
            "field": "highlightText",
            "usageId": "cmuygq4vf0180sg4czc6vz3ct",
            "company": {
              "name": "SynthLabs",
              "slug": "synthlabs"
            }
          }
        ],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 5,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published"
      },
      {
        "selectionBasis": "category",
        "does": "Cloud serverless pour l'inférence IA, l'entraînement sur GPU, les traitements batch et les sandboxes d'agents.",
        "name": "Modal",
        "slug": "modal",
        "companies": 43,
        "mix": {
          "confirmed": 1,
          "declared": 15,
          "observed": 27,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T13:46:20.168Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 34,
          "listed": [
            {
              "usageId": "cmuvwo3pp00d5sp1ox7xjhkkn",
              "company": "Doordash",
              "slug": "doordash",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-17T06:32:46.749Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwo3pp00d5sp1ox7xjhkkn",
                "method": "LOGO_WALL",
                "original": "https://modal.com/customers"
              }
            },
            {
              "usageId": "cmuvwo6s700desp1odtaxs5qa",
              "company": "Meta",
              "slug": "meta",
              "proof": "OBSERVED",
              "size": "MEGA",
              "lastCheckedAt": "2026-11-17T06:32:46.749Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwo6s700desp1odtaxs5qa",
                "method": "LOGO_WALL",
                "original": "https://modal.com/customers"
              }
            }
          ],
          "overflow": 32
        },
        "strongProofPeers": {
          "total": 16,
          "listed": [
            {
              "usageId": "cmuvwo79500dhsp1oye8xlnnu",
              "company": "Decagon",
              "slug": "decagon",
              "proof": "CONFIRMED",
              "size": "MEDIUM",
              "lastCheckedAt": "2026-11-17T09:51:27.258Z",
              "source": {
                "url": "https://www.usedby.ai/claims/cmuvwo79500dhsp1oye8xlnnu",
                "method": "CASE_STUDY",
                "original": "https://decagon.ai/blog/real-time-voice-ai-on-modal"
              }
            },
            {
              "usageId": "sp_usage_depot_modal_20261122_mwrtz5w4",
              "company": "Depot",
              "slug": "depot",
              "proof": "DECLARED",
              "size": null,
              "lastCheckedAt": "2026-11-22T13:46:20.168Z",
              "source": {
                "url": "https://www.usedby.ai/claims/sp_usage_depot_modal_20261122_mwrtz5w4",
                "method": "SUBPROCESSOR_LIST",
                "original": "https://depot.dev/legal/subprocessors"
              }
            },
            {
              "usageId": "sp_usage_rollbar_modal_20261122_mwrtjwix",
              "company": "Rollbar",
              "slug": "rollbar",
              "proof": "DECLARED",
              "size": null,
              "lastCheckedAt": "2026-11-22T13:34:28.412Z",
              "source": {
                "url": "https://www.usedby.ai/claims/sp_usage_rollbar_modal_20261122_mwrtjwix",
                "method": "SUBPROCESSOR_LIST",
                "original": "https://docs.rollbar.com/docs/data-subprocessors"
              }
            }
          ],
          "overflow": 13
        },
        "why": [],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 18,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published",
        "whyOverflow": 1
      },
      {
        "selectionBasis": "category",
        "does": "Plateforme cloud pour exécuter des IA open-source",
        "name": "Together AI",
        "slug": "together-ai",
        "companies": 39,
        "mix": {
          "confirmed": 20,
          "declared": 3,
          "observed": 16,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T13:39:31.032Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 34,
          "listed": [],
          "overflow": 34
        },
        "strongProofPeers": {
          "total": 23,
          "listed": [],
          "overflow": 23
        },
        "why": [],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 9,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published",
        "whyOverflow": 2
      },
      {
        "selectionBasis": "category",
        "does": "La plateforme pour développer l'IA générative à l'échelle de production",
        "name": "AWS Bedrock",
        "slug": "aws-bedrock",
        "companies": 37,
        "mix": {
          "confirmed": 16,
          "declared": 6,
          "observed": 8,
          "signal": 7
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T13:44:11.350Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 30,
          "listed": [],
          "overflow": 30
        },
        "strongProofPeers": {
          "total": 22,
          "listed": [],
          "overflow": 22
        },
        "why": [],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 12,
          "signalCompanies": 5,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published",
        "whyOverflow": 2
      },
      {
        "selectionBasis": "category",
        "does": "Systèmes IA à l'échelle du wafer et API d'inférence cloud qui sert des modèles ouverts et de pointe à très haute vitesse.",
        "name": "Cerebras",
        "slug": "cerebras",
        "companies": 37,
        "mix": {
          "confirmed": 0,
          "declared": 4,
          "observed": 33,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T12:56:33.870Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 34,
          "listed": [],
          "overflow": 34
        },
        "strongProofPeers": {
          "total": 4,
          "listed": [],
          "overflow": 4
        },
        "why": [],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 7,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "reason_not_published"
      },
      {
        "selectionBasis": "category",
        "does": "Cloud IA avec calcul GPU NVIDIA, stockage et Kubernetes managé pour entraîner et servir des modèles à grande échelle.",
        "name": "CoreWeave",
        "slug": "coreweave",
        "companies": 26,
        "mix": {
          "confirmed": 12,
          "declared": 4,
          "observed": 10,
          "signal": 0
        },
        "countryCount": null,
        "latestCheck": "2026-11-22T13:13:26.440Z",
        "alreadyUses": false,
        "largestPeers": {
          "total": 22,
          "listed": [],
          "overflow": 22
        },
        "strongProofPeers": {
          "total": 16,
          "listed": [],
          "overflow": 16
        },
        "why": [],
        "momentum": {
          "status": "available",
          "windowDays": 90,
          "companies": 7,
          "signalCompanies": 0,
          "basis": "public_proof_arrival"
        },
        "reasonStatus": "published",
        "whyOverflow": 2
      }
    ],
    "overflow": 4
  },
  "evidenceScope": "category",
  "categoryFallback": {
    "total": 0,
    "listed": [],
    "overflow": 0
  },
  "alreadyUses": [],
  "alreadyUsesTotal": 0,
  "alreadyUsesOverflow": 0,
  "summary": "80 entreprises sur usedby utilisent Baseten avec une preuve publique (ensemble du marché). T4; 26 CONFIRMED, 20 DECLARED, 34 OBSERVED, 0 SIGNAL. https://www.usedby.ai/fr/for?category=llm-inference",
  "compareUrl": "https://www.usedby.ai/fr/compare/baseten-vs-fireworks-ai",
  "limitations": [
    "Les usages publics sur usedby ne mesurent pas les parts de marché et ne constituent pas une recommandation. Les exigences ne sont pas évaluées.",
    "Les tailles inconnues sont exclues de la vue des plus grands pairs. Aucun usage n’est masqué en raison de son ancienneté.",
    "T1 est ignoré, car la taille de l’entreprise est inconnue.",
    "Le secteur est inconnu ; les données ne sont pas filtrées par secteur."
  ],
  "source": {
    "url": "https://www.usedby.ai/fr/for?category=llm-inference",
    "terms": "https://www.usedby.ai/terms",
    "citation": "Source: usedby, https://www.usedby.ai/fr/for?category=llm-inference"
  }
}
```
