Fireworks Fire Pass logo

Fireworks Fire Pass

LLM 4.0/10

GLM 5.2 Fast + Kimi K2.7 Code Fast with zero per-token charges (fair-use), 100+ tok/s first-party serving, 256K context; OpenAI- AND Anthropic-compatible endpoints (Claude Code, OpenCode, Cline, OpenClaw, custom agents)

Tokens / month
5B
Tokens / day
200M

Fireworks Fire Pass scores 4/10 for free value in LLM. The main constraint: Paid — $49/mo — INVITE-ONLY (v3 closed beta). GLM 5.2 Fast + Kimi K2.7 Code Fast with zero per-token charges (fair-use), 100+ tok/s first-party serving, 256K context; OpenAI- AND Anthropic-compatible endpoints (Claude Code, OpenCode, Cline, OpenClaw, custom agents).

Category LLM Kind Paid plan Caveats Invite-only since v2 sunset (auto-renew disabled Jun 6 2026; v3 public release TBD); non-production coding use only; soft fair-use cap — 429s and warning emails reported at ~100M+ tokens/day; one account per user

Free value score

4.0 / 10
Show the math
Model Quality 9.2/10
value = 51 artificial_analysis_score
weighted score = 0.2 * 9.2/10
Access 10/10
value = 10/10
weighted score = 0.1 * 10/10
Price Efficiency 1.6/10
value = 0.02 value/$
weighted score = 0.7 * 1.6/10

Key metrics

GLM-5.2 51 Best model
5.0B T/mo Free tokens
200M T/day Tokens/day
0.0204x Value ratio

What you get

Paid — $49/mo — INVITE-ONLY (v3 closed beta). GLM 5.2 Fast + Kimi K2.7 Code Fast with zero per-token charges (fair-use), 100+ tok/s first-party serving, 256K context; OpenAI- AND Anthropic-compatible endpoints (Claude Code, OpenCode, Cline, OpenClaw, custom agents)

Effectively-unlimited GLM 5.2 Fast + K2.7 Code Fast for $49; users report 85M-200M tokens/day, 4-5B tokens/mo sustained => 100x+ face value IF you can get an invite

Gotchas

No gotchas documented yet. Contribute one if you know a constraint we missed.

Sources

paid value llm-providers
Open service ← All LLM

Alternatives

Same category · similar quota
Kimi Code AllegrettoKimi Code Allegretto · paid · Paid plan
3.8

5x Moderato credits for Kimi K2.7 Code / HighSpeed / K2.6 + priority queue (4x speed) + Agent Swarm (50 uses); same OpenAI + Anthropic endpoints; community-consensus best-value Kimi tier — most heavy daily users never hit the weekly cap

0.03×42 AA0.0×
May be outdatedChecked 23d ago
Best valuePaid only
MiniMax Token Plan MaxMiniMax Token Plan Max · paid · Paid plan
3.7

~5.1B MiniMax-M3 tokens/mo; 4-5 concurrent agents; video gen 3 clips/day; same OpenAI + Anthropic endpoints; community reports sustaining ~1B tokens/day at high cache-hit

0.02×44.4 AA0.0×
May be outdatedChecked 23d ago
Best valueFlat ratePaid only
MiniMax Token Plan PlusMiniMax Token Plan Plus · paid · Paid plan
4.3

~1.7B MiniMax-M3 tokens/mo (M3 ~98 tok/s, 1M context) + M2.7/M2.7-highspeed + image/speech/music; 3-4 concurrent agents; OpenAI- AND Anthropic-compatible endpoints for Claude Code/OpenClaw/custom agents

0.05×44.4 AA0.1×
May be outdatedChecked 23d ago
Best valueFlat ratePaid only