# Fireworks AI vs Together AI

**Leader by vioscaleAI score:** Together AI

| Attribute | Fireworks AI | Together AI |
|---|---|---|
| **vioscaleAI score** | 61.1 (55% (medium)) | 76.7 (59% (medium)) |
| content.faq | `[{"answer":"Fireworks provides serverless and on-demand inference for open source models and custom trained versions, with tools for model fine-tuning, optimization, and deployment. It supports per-token and per-GPU billing with API compatibility for existing model clients. It is indexed under Inference Providers.","source":"https://fireworks.ai/pricing","question":"What is Fireworks AI?","confidence":0.6},{"answer":"Fireworks AI offers a free tier, so you can start without paying. Paid plans start at $0.50. Prices are published openly on the vendor's own pricing page. Pricing changes often, so verify at source before relying on it.","source":"https://fireworks.ai/pricing","question":"Is Fireworks AI free?","confidence":0.6},{"answer":"Fireworks AI supports the web and a command-line interface. Platforms we have not confirmed are simply not listed here rather than ruled out.","source":"https://fireworks.ai/pricing","question":"What platforms does Fireworks AI support?","confidence":0.6},{"answer":"We have only confirmed a cloud / SaaS deployment for Fireworks AI, so it appears to be vendor-hosted. If a self-hosted option exists we have not found it documented publicly.","source":"https://fireworks.ai/pricing","question":"Can Fireworks AI be self-hosted?","confidence":0.6},{"answer":"We have confirmed 6 integrations for Fireworks AI, including OpenCode, Cline, Aider, External APIs via function calling, OpenAI and Anthropic. This is what we could verify from public sources, so the vendor may support others we have not indexed.","source":"https://fireworks.ai/pricing","question":"What does Fireworks AI integrate with?","confidence":0.6},{"answer":"We have independently confirmed SOC 2, HIPAA and GDPR for Fireworks AI. Certifications we do not list are ones we have not been able to verify from public sources, which is not the same as Fireworks AI not holding them. Always confirm compliance directly before you rely on it.","source":"https://fireworks.ai/blog/fireworks-ai-achieves-soc-2-type-ii-and-hipaa-compliance","question":"What security certifications does Fireworks AI have?","confidence":0.7},{"answer":"Fireworks AI is available worldwide. Its primary market is the United States.","source":"https://fireworks.ai/pricing","question":"Where is Fireworks AI available?","confidence":0.75}]` | `[{"answer":"A cloud platform enabling developers to deploy, fine-tune, and run open-source AI models with optimized inference, serverless scaling, and dedicated computing resources. It is indexed under MLOps & LLMOps Tools.","source":"https://www.together.ai/blog/august-2023-pricing-update","question":"What is Together AI?","confidence":0.6},{"answer":"Together AI offers a free tier, so you can start without paying. Paid plans start at $0.00. Prices are published openly on the vendor's own pricing page. Pricing changes often, so verify at source before relying on it.","source":"https://www.together.ai/blog/august-2023-pricing-update","question":"Is Together AI free?","confidence":0.6},{"answer":"Together AI supports the web and a command-line interface. Platforms we have not confirmed are simply not listed here rather than ruled out.","source":"https://www.together.ai/blog/august-2023-pricing-update","question":"What platforms does Together AI support?","confidence":0.6},{"answer":"Yes. Together AI can be deployed cloud / SaaS and self-hosted, so it does not have to run on the vendor's infrastructure.","source":"https://www.together.ai/blog/august-2023-pricing-update","question":"Can Together AI be self-hosted?","confidence":0.6},{"answer":"We have independently confirmed SOC 2, ISO 27001 and GDPR for Together AI. Certifications we do not list are ones we have not been able to verify from public sources, which is not the same as Together AI not holding them. Always confirm compliance directly before you rely on it.","source":"https://trust.together.ai","question":"What security certifications does Together AI have?","confidence":0.7},{"answer":"Together AI is available worldwide. Its primary market is the United States.","source":"https://www.together.ai/blog/august-2023-pricing-update","question":"Where is Together AI available?","confidence":0.5}]` |
| deployment.options | `{"cloud":true,"hybrid":true,"on_prem":true}` | `{"cloud":true}` |
| description.long | Fireworks provides serverless and on-demand inference for open source models and custom trained versions, with tools for model fine-tuning, optimization, and deployment. It supports per-token and per-GPU billing with API compatibility for existing model clients. | Together AI provides cloud infrastructure for deploying, fine-tuning, and training open-source AI models. It offers flexible deployment options including serverless inference, batch processing, reserved throughput, and dedicated GPU instances with optimized performance and cost efficiency. |
| features.capabilities | `{"soc2_type_ii":true,"compute_model":"multi_tenant_hybrid","pricing_model":"per_million_tokens","openai_compatible_rest_api_schema":true,"bring_your_own_weights_byow_hosting":true,"dynamic_batching_and_kv_cache_management":true,"massive_context_window_support_100k_plus":true,"hipaa_baa_compliance_for_medical_inference":true,"multimodal_vision_and_audio_in_out_support":true,"native_function_calling_and_strict_json_mode":true,"streaming_server_sent_events_sse_token_yield":true,"lpu_or_custom_silicon_for_extreme_low_latency":true,"zero_data_retention_training_opt_out_enterprise":true}` | `{"role":"platform","evaluation":true,"soc2_type_ii":false,"compute_model":"multi_tenant_hybrid","managed_cloud":true,"model_serving":true,"pricing_model":"per_million_tokens","self_hostable":false,"multi_provider":true,"vpc_deployment":true,"otel_compatible":false,"no_train_on_data":"yes","llm_observability":false,"prompt_management":false,"framework_agnostic":true,"experiment_tracking":false,"openai_compatible_rest_api_schema":true,"bring_your_own_weights_byow_hosting":true,"dynamic_batching_and_kv_cache_management":false,"massive_context_window_support_100k_plus":true,"hipaa_baa_compliance_for_medical_inference":false,"multimodal_vision_and_audio_in_out_support":true,"native_function_calling_and_strict_json_mode":true,"streaming_server_sent_events_sse_token_yield":false,"lpu_or_custom_silicon_for_extreme_low_latency":false,"zero_data_retention_training_opt_out_enterprise":true}` |
| integrations.count | 2 | - |
| integrations.list | `[{"name":"OpenCode"},{"name":"Cline"},{"name":"Aider"},{"name":"External APIs via function calling"},{"name":"OpenAI","native":true,"category":"api_compatibility","direction":"bidirectional"},{"name":"Anthropic","native":true,"category":"api_compatibility","direction":"bidirectional"}]` | - |
| market.availability | `{"primaryMarkets":["US"],"availabilityScope":"global","availableCountries":[],"notAvailableCountries":[]}` | `{"primaryMarkets":["US"],"availabilityScope":"global","availableCountries":[],"notAvailableCountries":[]}` |
| platform.support | `{"cli":true,"web":true}` | `{"cli":true,"web":true}` |
| pricing | `{"type":"usage","plans":[{"free":false,"name":"Serverless Inference","summary":"Usage-based per-token pricing","features":["Pay per input token","Pay per output token","Pay per cached token","OpenAI and Anthropic compatible API","Priority and Fast serving options"],"commitment":"monthly","description":"Per-token usage-based billing with pre-paid credits and no cold starts","contactSales":false},{"free":false,"name":"Embeddings","summary":"$0.008–$0.1 per 1M input tokens","features":["Multiple embedding models","Tiered by model size"],"commitment":"monthly","components":[{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":0.008,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":0.016,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":0.1,"currency":"USD"}],"description":"Embedding model inference with tiered pricing by model size","contactSales":false},{"free":false,"name":"Managed Training","summary":"$0.50–$12 per 1M training tokens depending on model size and method","features":["LoRA SFT","LoRA DPO","Full Parameter SFT","Full Parameter DPO","Quantization-aware training"],"commitment":"monthly","components":[{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":0.5,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":1,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":2,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":3,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":6,"currency":"USD"},{"per":{"qty":1000000,"unit":"tokens"},"kind":"metered","amount":12,"currency":"USD"}],"description":"Supervised and preference fine-tuning with pricing based on model size and method","contactSales":false},{"free":false,"name":"Serverless Training API","summary":"Usage-based pricing per 1M tokens for prefill, cached prefill, sample, and train","features":["Shared trainer pool","No provisioning required","No idle costs","Cached prefill option"],"commitment":"monthly","description":"LoRA training with usage-based per-token pricing on a shared trainer pool","contactSales":false},{"free":false,"name":"On-Demand Deployments","summary":"$0.134–$0.334 per minute for dedicated GPU deployments","features":["H100, B200, B300, GB300 GPU options","Multi-region support","Post-trained model support"],"components":[{"per":{"qty":1,"unit":"minute"},"kind":"metered","amount":0.134,"currency":"USD"},{"per":{"qty":1,"unit":"minute"},"kind":"metered","amount":0.217,"currency":"USD"},{"per":{"qty":1,"unit":"minute"},"kind":"metered","amount":0.25,"currency":"USD"},{"per":{"qty":1,"unit":"minute"},"kind":"metered","amount":0.334,"currency":"USD"}],"description":"Dedicated GPU instances billed per minute or hour for custom workloads","contactSales":true}],"summary":"Usage-based pay-per-token for serverless; managed training at $0.50–$12/1M tokens; dedicated GPU from $0.134–$0.334/minute","currency":"USD","freeTier":false,"sourceUrl":"https://fireworks.ai/pricing","retrievedAt":"2026-10-02T16:20:24.427Z","startingPrice":{"unit":"tokens","amount":0.008,"currency":"USD"},"billingPeriods":["month"]}` | `{"type":"usage","plans":[{"free":true,"name":"Build","summary":"Free tier","features":["Discord community support"],"description":"Community tier","contactSales":false},{"free":false,"name":"Scale","summary":"From $0.0015 per 1M tokens","features":["Serverless inference","Batch API","Standard Support (business hours)"],"components":[{"per":{"qty":1,"unit":"million tokens"},"kind":"metered","amount":0.0015,"currency":"USD"}],"description":"Pay-per-token inference with Standard Support","contactSales":false},{"free":false,"name":"Enterprise","summary":"Contact sales for reserved capacity pricing","features":["Provisioned throughput","Dedicated GPU instances","Silver/Gold Support","Dedicated support engineer"],"description":"Reserved capacity and dedicated infrastructure with Silver/Gold Support","contactSales":true}],"summary":"Pay-per-token usage starting from $0.0015 per 1M tokens. Free tier available. Dedicated instances and reserved throughput available upon request.","currency":"USD","freeTier":true,"sourceUrl":"https://www.together.ai/pricing","retrievedAt":"2026-10-02T16:26:49.735Z","startingPrice":{"unit":"million tokens","amount":0.0015,"currency":"USD"}}` |
| pricing.free_tier | no | yes |
| pricing.model | commercial | freemium |
| pricing.price_level | low | low |
| pricing.starting_price | `{"amount":0.008,"currency":"USD"}` | `{"amount":0.0015,"currency":"USD"}` |
| pricing.transparent | yes | yes |
| reliability.sla_pct | - | 99 |
| reliability.status_page | - | yes |
| security.certifications | `[{"name":"SOC 2 Type II","level":"Type II","status":"active","auditor":"Vanta & Advantage Partners","evidenceUrl":"https://fireworks.ai/blog/fireworks-ai-achieves-soc-2-type-ii-and-hipaa-compliance"},{"name":"HIPAA","status":"active","auditor":"Vanta & Advantage Partners","evidenceUrl":"https://fireworks.ai/blog/fireworks-ai-achieves-soc-2-type-ii-and-hipaa-compliance"}]` | - |
| security.disclosure_policy | yes | yes |
| security.gdpr | yes | yes |
| security.hipaa | yes | - |
| security.iso27001 | - | yes |
| security.soc2 | yes | yes |
| security.trust_center | https://fireworks.ai/blog/fireworks-ai-achieves-soc-2-type-ii-and-hipaa-compliance | https://www.together.ai/privacy |

## Capabilities (Inference Providers)

| Capability | Fireworks AI | Together AI |
|---|:--:|:--:|
| **Capabilities** |  |  |
| Compute model | Multi tenant hybrid | Multi tenant hybrid |
| Zero data retention training opt out enterprise | ✓ | ✓ |
| Openai compatible rest API schema | ✓ | ✓ |
| Native function calling and strict json mode | ✓ | ✓ |
| Multimodal vision and audio in out support | ✓ | ✓ |
| Streaming server sent events sse token yield | ✓ | ✗ |
| Massive context window support 100k plus | ✓ | ✓ |
| Lpu or custom silicon for extreme low latency | ✓ | ✗ |
| Bring your own weights byow hosting | ✓ | ✓ |
| Dynamic batching and kv cache management | ✓ | ✗ |
| SOC2 type ii | ✓ | ✗ |
| HIPAA baa compliance for medical inference | ✓ | ✗ |
| Pricing model | Per million tokens | Per million tokens |

*Source: vioscaleAI. Generated 2026-10-16T23:30:46.415Z. "-" = undocumented, not absent.*