← variables.md

2026-10-15

143 change(s) in the model market

model price changes

• ~deepseek/deepseek-flash-latest: prompt price $0.02/M → $0.024/M (+17.9%)
• ~deepseek/deepseek-flash-latest: completion price $0.891/M → $0.72/M (−19.2%)
• ~google/gemini-pro-latest: completion price $12/M → $10/M (−16.7%)
• ~moonshotai/kimi-latest: prompt price $0.496/M → $0.3/M (−39.5%)
• ~z-ai/glm-latest: prompt price $0.058/M → $0.024/M (−58.3%)
• deepseek/deepseek-v4-pro-0813: prompt price $0.66/M → $1.32/M (+100.0%)
• deepseek/deepseek-v4-pro-0813: completion price $1.98/M → $3.96/M (+100.0%)
• deepseek/deepseek-v4.1-flash: prompt price $0.02/M → $0.024/M (+17.9%)
• deepseek/deepseek-v4.1-flash: completion price $0.891/M → $0.72/M (−19.2%)
• google/gemma-4-31b-it: prompt price $0.09/M → $0.2/M (+122.2%)
• google/gemma-4-31b-it: completion price $0.34/M → $0.4/M (+17.6%)
• minimax/minimax-m2.5: completion price $1.08/M → $0.95/M (−12.0%)
• mistralai/mistral-small-3.2-24b-instruct: prompt price $0.094/M → $0.075/M (−20.0%)
• mistralai/mistral-small-3.2-24b-instruct: completion price $0.25/M → $0.2/M (−20.0%)
• moonshotai/kimi-k3: prompt price $0.62/M → $0.45/M (−27.4%)
• openai/gpt-5.6-sol: prompt price $2/M → $4/M (+100.0%)
• openai/gpt-5.6-sol: completion price $10/M → $20/M (+100.0%)
• openai/gpt-5.6-sol-pro: prompt price $2/M → $4/M (+100.0%)
• openai/gpt-5.6-sol-pro: completion price $10/M → $20/M (+100.0%)
• openai/gpt-5.6-sol-pro:batch: prompt price $1/M → $2/M (+100.0%)
• openai/gpt-5.6-sol-pro:batch: completion price $5/M → $10/M (+100.0%)
• openai/gpt-5.6-sol:batch: prompt price $1/M → $2/M (+100.0%)
• openai/gpt-5.6-sol:batch: completion price $5/M → $10/M (+100.0%)
• openai/gpt-oss-120b: prompt price $0.037/M → $0.15/M (+305.4%)
• openai/gpt-oss-120b: completion price $0.17/M → $0.6/M (+252.9%)
• qwen/qwen3.6-35b-a3b: prompt price $0.1/M → $0.15/M (+50.0%)
• qwen/qwen3.6-35b-a3b: completion price $0.95/M → $1/M (+5.3%)
• qwen/qwen3.8-27b: prompt price $0.04/M → $0.24/M (+500.0%)
• qwen/qwen3.8-27b: completion price $2.3/M → $2.2/M (−4.3%)
• z-ai/glm-5: prompt price $0.6/M → $1/M (+66.7%)
• z-ai/glm-5: completion price $1.92/M → $3.2/M (+66.7%)
• z-ai/glm-5.1: prompt price $1.33/M → $0.966/M (−27.4%)
• z-ai/glm-5.1: completion price $4.18/M → $3.04/M (−27.4%)
• z-ai/glm-5.2: prompt price $0.05/M → $0.02/M (−60.0%)
• z-ai/glm-5.3: prompt price $0.079/M → $0.03/M (−62.0%)
• z-ai/glm-5.3-flash: prompt price $0.036/M → $0.15/M (+316.7%)
• z-ai/glm-5.3-flash: completion price $0.9/M → $0.5/M (−44.4%)

model context changes

• stepfun/step-5-preview: context length 1,000,000 → 1,048,576

quantization changes

• DeepInfra: tencent/hy4-preview quantization fp8 → fp4

provider price changes

• Wafer: deepseek/deepseek-v4-flash prompt price $0.4/M → $0.42/M (+5.0%)
• Phala: deepseek/deepseek-v4-flash-0731 prompt price $0.308/M → $0.44/M (+42.9%)
• Phala: deepseek/deepseek-v4-flash-0731 completion price $0.924/M → $1.32/M (+42.9%)
• Alibaba: deepseek/deepseek-v4-pro prompt price $1.56/M → $1.42/M (−9.2%)
• Alibaba: deepseek/deepseek-v4-pro completion price $3.12/M → $2.83/M (−9.2%)
• Reka: deepseek/deepseek-v4-pro completion price $10.5/M → $2.1/M (−80.0%)
• DeepSeek: deepseek/deepseek-v4-pro-0813 prompt price $0.66/M → $1.32/M (+100.0%)
• DeepSeek: deepseek/deepseek-v4-pro-0813 completion price $1.98/M → $3.96/M (+100.0%)
• Phala: deepseek/deepseek-v4-pro-0813 prompt price $0.957/M → $1.45/M (+51.5%)
• Phala: deepseek/deepseek-v4-pro-0813 completion price $2.88/M → $4.36/M (+51.5%)
• Sail Research: deepseek/deepseek-v4-pro-0813 prompt price $0.48/M → $0.896/M (+86.6%)
• Sail Research: deepseek/deepseek-v4-pro-0813 completion price $1.98/M → $0.99/M (−50.0%)
• Sail Research: deepseek/deepseek-v4-pro-0813 prompt price $0.48/M → $0.896/M (+86.6%)
• Sail Research: deepseek/deepseek-v4-pro-0813 completion price $1.98/M → $0.99/M (−50.0%)
• Baidu: deepseek/deepseek-v4.1-flash prompt price $0.171/M → $0.3/M (+75.4%)
• Baidu: deepseek/deepseek-v4.1-flash completion price $0.684/M → $1.2/M (+75.4%)
• CoreWeave: deepseek/deepseek-v4.1-flash prompt price $0.2/M → $0.15/M (−25.0%)
• DeepSeek: deepseek/deepseek-v4.1-flash prompt price $0.15/M → $0.3/M (+100.0%)
• DeepSeek: deepseek/deepseek-v4.1-flash completion price $0.6/M → $1.2/M (+100.0%)
• DigitalOcean: deepseek/deepseek-v4.1-flash prompt price $0.165/M → $0.225/M (+36.4%)
• DigitalOcean: deepseek/deepseek-v4.1-flash completion price $0.66/M → $0.9/M (+36.4%)
• Morph: deepseek/deepseek-v4.1-flash prompt price $0.028/M → $0.033/M (+17.9%)
• Morph: deepseek/deepseek-v4.1-flash completion price $0.8/M → $0.6/M (−25.0%)
• OpenInference: deepseek/deepseek-v4.1-flash prompt price $0.02/M → $0.024/M (+17.9%)
• OpenInference: deepseek/deepseek-v4.1-flash completion price $0.891/M → $0.72/M (−19.2%)
• Relace: deepseek/deepseek-v4.1-flash prompt price $0.022/M → $0.026/M (+17.9%)
• Relace: deepseek/deepseek-v4.1-flash completion price $0.6/M → $0.8/M (+33.3%)
• Wafer: deepseek/deepseek-v4.1-flash completion price $0.99/M → $0.65/M (−34.3%)
• NextBit: google/gemma-3-27b-it prompt price $0.07/M → $0.08/M (+14.3%)
• NextBit: google/gemma-3-27b-it completion price $0.275/M → $0.35/M (+27.3%)
• Decart: moonshotai/kimi-k3 prompt price $2.34/M → $2.13/M (−9.0%)
• Decart: moonshotai/kimi-k3 completion price $11.7/M → $10.7/M (−9.0%)
• InferenceNet: moonshotai/kimi-k3 prompt price $0.62/M → $0.45/M (−27.4%)
• InferenceNet: moonshotai/kimi-k3 prompt price $2.99/M → $2.49/M (−16.7%)
• InferenceNet: moonshotai/kimi-k3 completion price $18/M → $15/M (−16.7%)
• Morph: moonshotai/kimi-k3 prompt price $0.67/M → $0.8/M (+19.4%)
• Morph: moonshotai/kimi-k3 completion price $20/M → $14.9/M (−25.5%)
• Relace: moonshotai/kimi-k3 prompt price $0.496/M → $0.32/M (−35.5%)
• Wafer: moonshotai/kimi-k3 prompt price $0.7/M → $0.33/M (−52.9%)
• Wafer: moonshotai/kimi-k3 completion price $10/M → $14.9/M (+49.0%)
• Wafer: moonshotai/kimi-k3 prompt price $0.72/M → $0.31/M (−56.9%)
• OpenAI: openai/gpt-5.6-sol prompt price $1/M → $2/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol completion price $5/M → $10/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol prompt price $2/M → $4/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol completion price $10/M → $20/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol prompt price $4/M → $8/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol completion price $20/M → $40/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol-pro prompt price $1/M → $2/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol-pro completion price $5/M → $10/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol-pro prompt price $2/M → $4/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol-pro completion price $10/M → $20/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol-pro prompt price $4/M → $8/M (+100.0%)
• OpenAI: openai/gpt-5.6-sol-pro completion price $20/M → $40/M (+100.0%)
• Darkbloom: qwen/qwen3.5-35b-a3b prompt price $0.08/M → $0.065/M (−18.8%)
• DeepInfra: xiaomi/mimo-v2.6-flash prompt price $0.14/M → $0.133/M (−5.0%)
• DeepInfra: xiaomi/mimo-v2.6-flash completion price $0.28/M → $0.266/M (−5.0%)
• Alibaba: z-ai/glm-5.2 prompt price $1.12/M → $0.966/M (−13.8%)
• Alibaba: z-ai/glm-5.2 completion price $3.52/M → $3.04/M (−13.8%)
• CoreWeave: z-ai/glm-5.2 prompt price $0.76/M → $0.62/M (−18.4%)
• StreamLake: z-ai/glm-5.2 prompt price $0.643/M → $0.613/M (−4.6%)
• StreamLake: z-ai/glm-5.2 completion price $2.02/M → $1.93/M (−4.6%)
• Wafer: z-ai/glm-5.2 prompt price $0.05/M → $0.02/M (−60.0%)
• Decart: z-ai/glm-5.3 prompt price $0.84/M → $0.92/M (+9.5%)
• Decart: z-ai/glm-5.3 completion price $2.65/M → $2.93/M (+10.6%)
• InferenceNet: z-ai/glm-5.3 completion price $4.4/M → $4.8/M (+9.1%)
• Mistral: z-ai/glm-5.3 prompt price $1.54/M → $1.4/M (−9.1%)
• Mistral: z-ai/glm-5.3 completion price $4.84/M → $4.4/M (−9.1%)
• Morph: z-ai/glm-5.3 prompt price $0.074/M → $0.092/M (+24.3%)
• Relace: z-ai/glm-5.3 prompt price $0.058/M → $0.024/M (−58.3%)
• Wafer: z-ai/glm-5.3 prompt price $0.17/M → $0.03/M (−82.4%)
• Inceptron: z-ai/glm-5.3-flash prompt price $0.105/M → $0.108/M (+2.6%)
• InferenceNet: z-ai/glm-5.3-flash prompt price $0.1/M → $0.083/M (−17.0%)
• Morph: z-ai/glm-5.3-flash prompt price $0.044/M → $0.069/M (+56.8%)
• Wafer: z-ai/glm-5.3-flash prompt price $0.095/M → $0.12/M (+26.3%)

provider context changes

• InferenceNet: moonshotai/kimi-k3 context length 250,000 → 1,048,576

providers added

• Alibaba now serves deepseek/deepseek-v4-flash (fp8, 1,000,000 ctx, $0.15/M in / $0.3/M out)
• BaseTen now serves deepseek/deepseek-v4-flash-0731 (fp8, 1,048,576 ctx, $0.13/M in / $0.26/M out)
• Alibaba now serves deepseek/deepseek-v4-pro (fp8, 1,000,000 ctx, $1.56/M in / $3.12/M out)
• BaseTen now serves deepseek/deepseek-v4.1-flash (fp32, 1,048,576 ctx, $0.3/M in / $1.2/M out)
• PrimeIntellect now serves deepseek/deepseek-v4.1-flash (unknown, 1,048,576 ctx, $0.3/M in / $1.2/M out)
• Google now serves google/gemini-4-argon (unknown, 1,048,576 ctx, $2.2/M in / $11/M out)
• Parasail now serves google/gemma-3-4b-it (fp8, 131,072 ctx, $0.04/M in / $0.05/M out)
• DeepInfra now serves inclusionai/ling-3.1-flash (fp4, 262,144 ctx, $0.3/M in / $0.9/M out)
• BaseTen now serves nvidia/nemotron-3-ultra-550b-a55b (fp4, 202,800 ctx, $0.6/M in / $2.4/M out)
• BaseTen now serves openai/gpt-oss-120b (fp4, 128,072 ctx, $0.1/M in / $0.5/M out)
• Reka now serves qwen/qwen3.5-27b (unknown, 32,768 ctx, $0.26/M in / $2.6/M out)
• Tenstorrent now serves qwen/qwen3.8-27b (fp4, 262,144 ctx, $0.02/M in / $0.5/M out)
• DeepInfra now serves stepfun/step-5-preview (fp8, 1,048,576 ctx, $1/M in / $2.7/M out)
• Alibaba now serves z-ai/glm-5.2 (fp8, 1,000,000 ctx, $1.12/M in / $3.52/M out)
• BaseTen now serves z-ai/glm-5.2 (fp8, 1,048,576 ctx, $1.4/M in / $4.4/M out)
• BaseTen now serves z-ai/glm-5.2 (fp8, 1,048,576 ctx, $2.1/M in / $6.6/M out)
• Relace now serves z-ai/glm-5.2 (unknown, 1,048,576 ctx, $0.019/M in / $4/M out)
• Alibaba now serves z-ai/glm-5.3 (unknown, 1,000,000 ctx, $2.8/M in / $8.8/M out)
• BaseTen now serves z-ai/glm-5.3 (fp4, 1,048,576 ctx, $1.4/M in / $4.4/M out)
• BaseTen now serves z-ai/glm-5.3 (fp8, 1,048,576 ctx, $2.1/M in / $6.6/M out)
• Mistral now serves z-ai/glm-5.3 (nvfp4, 1,048,576 ctx, $1.54/M in / $4.84/M out)
• Tenstorrent now serves z-ai/glm-5.3 (fp8, 1,044,480 ctx, $0.6/M in / $2.5/M out)
• BaseTen now serves z-ai/glm-5.3-flash (fp8, 1,048,576 ctx, $0.15/M in / $0.5/M out)
• NextBit now serves z-ai/glm-5.3-flash (fp4, 1,048,576 ctx, $0.19/M in / $0.6/M out)

providers removed

• Alibaba no longer serves deepseek/deepseek-v4-flash (was fp8)
• Amazon Bedrock no longer serves qwen/qwen3-coder-30b-a3b-instruct (was unknown)
• Decart no longer serves z-ai/glm-5.2 (was mxfp4)
• InferenceNet no longer serves z-ai/glm-5.2 (was fp4)
• Alibaba no longer serves z-ai/glm-5.3 (was unknown)

openrouter catalog snapshots 2026-10-14 → 2026-10-15 · schema 1 · generated 2026-10-15T09:55:33.222Z