Fireworks AI vs Together AI
What the benchmark found. Fireworks AI had the lower end-to-end p99 latency at 10.49 seconds. Fireworks AI had the higher task success rate at 99.5%. Both figures come from 600 submitted requests per provider. Measured across 10 LLM inference providers serving GLM 5.3 Flash, 600 requests each, Sep 3, 2026.
Providers measured. Baseten, DeepInfra, Fireworks AI, Modal, Nebius, Novita AI, Parasail, Telnyx, Together AI, and Z.AI, each serving GLM 5.3 Flash behind an OpenAI-compatible chat-completions endpoint.
What this page compares. Fireworks AI and Together AI expose different model identifiers but were held to one logical inference contract. This direct comparison shows where each led and the size of the measured gap.
How to read the result. Both providers received identical selected items in the same temporal rounds. Each axis below is computed independently; the faster complete response is not automatically the more accurate or reliable API.
Workload and configuration. Every provider received the same 600 deliberately easy structured tasks: 200 meeting-note lookups, 200 ticket-triage classifications, and 200 contract-term extractions. GLM 5.3 Flash ran with streaming enabled, reasoning low, temperature 0, top-p 1, a 256-token output ceiling, one attempt, a 20-second timeout, and per-provider concurrency one.
APIs used. The runner called each provider's OpenAI-compatible chat-completions surface with the model identifiers below; the links open the provider documentation used to verify each adapter.
- Fireworks AI:
accounts/fireworks/models/glm-5p3-flash· API docs ↗ - Together AI:
zai-org/GLM-5.3-Flash· API docs ↗
Fireworks AI vs Together AI: which is faster, and which is more accurate?
Lower wins for latency and failure rate; higher wins for tokens per second and task success. The gap is shown in the native unit of each metric.
| Metric | Fireworks AI | Together AI | Measured leader | Absolute gap |
|---|---|---|---|---|
| End-to-end p99 | 10.5 s | 12.4 s | Fireworks AI | 1962 ms |
| End-to-end p95 | 6.72 s | 3.66 s | Together AI | 3053 ms |
| Time to first token p95 | 5.83 s | 2.26 s | Together AI | 3565 ms |
| Tokens per second p50 | 58.2 tok/s | 151.6 tok/s | Together AI | 93.4 tok/s |
| Tokens per second p95 | 27.9 tok/s | 34.0 tok/s | Together AI | 6.2 tok/s |
| Task success | 99.5% | 98.5% | Fireworks AI | 1.0 pp |
| Failure rate | 0.0% | 1.0% | Fireworks AI | 1.0 pp |
Provider by provider, in the order above
- Fireworks AI: 1.95 seconds median, 10.49 seconds end-to-end p99, time to first token p95 5.83 seconds, 58.2 tokens per second, 99.5% task success, 0.0% failures.
- Together AI: 609 ms median, 12.45 seconds end-to-end p99, time to first token p95 2.26 seconds, 151.6 tokens per second, 98.5% task success, 1.0% failures.
Where should I compare Fireworks AI with Together AI?
- Schema-constrained API responses. Start with task success, then compare end-to-end p95 and p99 because downstream code waits for the complete structured result.
- Streamed copilot output. Compare time to first token p95 for response start and tokens per second p95 for the slow throughput tail.
- Latency-sensitive extraction. Use end-to-end p99 with failure rate when this model call sits directly on a user-visible critical path.
How is LLM inference provider latency and accuracy measured here?
- Same model. Every endpoint serves GLM 5.3 Flash; the provider layer is the only variable.
- Same request. One-turn, non-tool chat completion in JSON-object mode, streaming on, temperature 0, top-p 1, reasoning low, 256-token ceiling, 20-second timeout, concurrency one.
- Client clock, pinned tokenizer. Time to first token, end-to-end latency and tokens per second come from the client; visible answer tokens are counted with a pinned
tiktoken o200k_basetokenizer, never provider-reported usage. - Exact task success. An answer counts only when it is schema-valid and exactly correct under the task rubric. Timeouts, HTTP errors and malformed responses stay in the denominator.
- Serverless endpoints only. Dedicated capacity is outside the comparison. Runner and per-request records: openbenchmarks-labs/inference ↗.
Fireworks AI vs Together AI: direct questions
Fireworks AI vs Together AI: who led on end-to-end p99?
Fireworks AI had the lower end-to-end p99 latency at 10.49 seconds. Fireworks AI had the higher task success rate at 99.5%. Both figures come from 600 submitted requests per provider.
Which API had the stronger slow-tail tokens per second?
Together AI had the stronger p95 tokens per second floor at 34.0 tokens per second. That is the throughput reached by at least 95% of measured responses.
Were Fireworks and Together evaluated with identical reasoning settings?
Yes. Both providers received the same logical low-reasoning setting, plus identical streaming, sampling, output-limit, task, and concurrency settings. Their provider-specific model identifiers are linked in the hero.
Same run, other questions
Read the complete LLM inference provider benchmark
Metric definitions, request scheduling, task rubrics, failure handling, and the complete provider scorecard live on the LLM Inference Provider Benchmark →