chore(pricing): Update fireworks-ai pricing

[siddharthsambharia-portkey]

🔄 Pricing Update: fireworks-ai

📊 Summary (complete_diff mode)

Change Type	Count
➕ Models added	16
🔄 Models updated (merged)	7

➕ New Models

• deepseek-v3p1
• deepseek-v3p2
• glm-4p7
• glm-5p1
• gpt-oss-120b
• gpt-oss-20b
• qwen3-vl-30b-a3b-instruct
• qwen3-vl-30b-a3b-thinking
• llama-v3p3-70b-instruct
• qwen3-8b
• qwen3p6-plus
• flux-1-dev-fp8
• flux-1-schnell-fp8
• flux-kontext-pro
• flux-kontext-max
• qwen3-embedding-8b

🔄 Updated Models

• glm-5
• kimi-k2-instruct-0905
• kimi-k2p5
• kimi-k2-thinking
• minimax-m2p1
• minimax-m2p5
• mixtral-8x22b-instruct

Model → Pricing Mapping

Model ID	Pricing Row	Input $/1M	Output $/1M	Cache $/1M
deepseek-v3p1	DeepSeek V3 Family (named)	$0.56	$1.68	$0.28 (50%)
deepseek-v3p2	DeepSeek V3 Family (named)	$0.56	$1.68	$0.28 (50%)
glm-4p7	GLM-4.7 (named)	$0.60	$2.20	$0.30 (50%)
glm-5	GLM-5 (named)	$1.00	$3.20	$0.20 (explicit)
glm-5p1	GLM-5.1 (named)	$1.40	$4.40	$0.26 (explicit)
kimi-k2-instruct-0905	Kimi K2 Instruct (named)	$0.60	$2.50	$0.30 (50%)
kimi-k2p5	Kimi K2.5 (named)	$0.60	$3.00	$0.10 (explicit)
kimi-k2-thinking	Kimi K2 Thinking (named)	$0.60	$2.50	$0.30 (50%)
minimax-m2p1	MiniMax M2 Family (named)	$0.30	$1.20	$0.03 (explicit)
minimax-m2p5	MiniMax M2 Family (named)	$0.30	$1.20	$0.03 (explicit)
gpt-oss-120b	OpenAI gpt-oss-120b (named)	$0.15	$0.60	$0.075 (50%)
gpt-oss-20b	OpenAI gpt-oss-20b (named)	$0.07	$0.30	$0.035 (50%)
qwen3-vl-30b-a3b-instruct	Qwen3 VL 30B A3B (named)	$0.15	$0.60	$0.075 (50%)
qwen3-vl-30b-a3b-thinking	Qwen3 VL 30B A3B (named)	$0.15	$0.60	$0.075 (50%)
llama-v3p3-70b-instruct	>16B tier	$0.90	$0.90	$0.45 (50%)
mixtral-8x22b-instruct	MoE 56.1B–176B tier	$1.20	$1.20	$0.60 (50%)
qwen3-8b	4B–16B tier	$0.20	$0.20	$0.10 (50%)
qwen3p6-plus	>16B tier (no named row found)	$0.90	$0.90	$0.45 (50%)
flux-1-dev-fp8	Image: FLUX.1 dev	$0.0005/step	—	—
flux-1-schnell-fp8	Image: FLUX.1 schnell	$0.00035/step	—	—
flux-kontext-pro	Image: FLUX Kontext Pro	$0.04/image	—	—
flux-kontext-max	Image: FLUX Kontext Max	$0.08/image	—	—
qwen3-embedding-8b	Embedding: Qwen3 8B	$0.10/1M	$0	—

Skipped: qwen3-reranker-8b (reranker — excluded per rules)

Batch pricing: 50% of serverless input/output for all text/vision models.
Cache pricing: 50% of input for most; explicit values used for GLM-5, GLM-5.1, Kimi K2.5, MiniMax M2.
Sources: Fireworks AI Models API + https://fireworks.ai/pricing (scraped April 10, 2026)

---

Generated by Pricing Agent on 2026-04-10