fix: add pricing for Cerebras models

- Llama 3.3 70B: /bin/sh.85 input / .20 output per M tokens
- Qwen 3 32B: /bin/sh.40 input / /bin/sh.80 output per M tokens
- Qwen 3 235B: /bin/sh.60 input / .20 output per M tokens
- Qwen 3 235B Instruct: /bin/sh.60 input / .20 output per M tokens
This commit is contained in:
Daniel Riccio 2025-07-29 20:12:21 -05:00
parent f9f619bf1f
commit c6eb4e889c
No known key found for this signature in database
GPG key ID: FFD5FD825F8E8209

View file

@ -11,8 +11,8 @@ export const cerebrasModels = {
contextWindow: 64000,
supportsImages: false,
supportsPromptCache: false,
inputPrice: 0,
outputPrice: 0,
inputPrice: 0.85,
outputPrice: 1.2,
description: "Smart model with ~2600 tokens/s",
},
"qwen-3-32b": {
@ -20,8 +20,8 @@ export const cerebrasModels = {
contextWindow: 64000,
supportsImages: false,
supportsPromptCache: false,
inputPrice: 0,
outputPrice: 0,
inputPrice: 0.4,
outputPrice: 0.8,
description: "SOTA coding performance with ~2500 tokens/s",
},
"qwen-3-235b-a22b": {
@ -29,8 +29,8 @@ export const cerebrasModels = {
contextWindow: 40000,
supportsImages: false,
supportsPromptCache: false,
inputPrice: 0,
outputPrice: 0,
inputPrice: 0.6,
outputPrice: 1.2,
description: "SOTA performance with ~1400 tokens/s",
},
"qwen-3-235b-a22b-instruct-2507": {
@ -38,8 +38,8 @@ export const cerebrasModels = {
contextWindow: 64000,
supportsImages: false,
supportsPromptCache: false,
inputPrice: 0,
outputPrice: 0,
inputPrice: 0.6,
outputPrice: 1.2,
description: "SOTA performance with ~1400 tokens/s",
supportsReasoningEffort: true,
},