@botpress/cognitive 0.1.42 → 0.1.43
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.turbo/turbo-build.log +7 -7
- package/dist/index.cjs +124 -34
- package/dist/index.cjs.map +2 -2
- package/dist/index.d.ts +1 -1
- package/dist/index.mjs +124 -34
- package/dist/index.mjs.map +2 -2
- package/package.json +1 -1
package/.turbo/turbo-build.log
CHANGED
|
@@ -1,28 +1,28 @@
|
|
|
1
1
|
|
|
2
|
-
> @botpress/cognitive@0.1.
|
|
2
|
+
> @botpress/cognitive@0.1.43 build /home/runner/work/botpress/botpress/packages/cognitive
|
|
3
3
|
> pnpm build:type && pnpm build:neutral && size-limit
|
|
4
4
|
|
|
5
5
|
|
|
6
|
-
> @botpress/cognitive@0.1.
|
|
6
|
+
> @botpress/cognitive@0.1.43 build:type /home/runner/work/botpress/botpress/packages/cognitive
|
|
7
7
|
> tsup --tsconfig tsconfig.build.json ./src/index.ts --dts-resolve --dts-only --clean
|
|
8
8
|
|
|
9
9
|
CLI Building entry: ./src/index.ts
|
|
10
10
|
CLI Using tsconfig: tsconfig.build.json
|
|
11
11
|
CLI tsup v8.0.2
|
|
12
12
|
DTS Build start
|
|
13
|
-
DTS ⚡️ Build success in
|
|
14
|
-
DTS dist/index.d.ts 625.
|
|
13
|
+
DTS ⚡️ Build success in 9146ms
|
|
14
|
+
DTS dist/index.d.ts 625.78 KB
|
|
15
15
|
|
|
16
|
-
> @botpress/cognitive@0.1.
|
|
16
|
+
> @botpress/cognitive@0.1.43 build:neutral /home/runner/work/botpress/botpress/packages/cognitive
|
|
17
17
|
> ts-node -T ./build.ts --neutral
|
|
18
18
|
|
|
19
19
|
Done
|
|
20
20
|
|
|
21
21
|
[1mdist/index.cjs[22m
|
|
22
22
|
Size limit: [32m[1m50 kB[22m[39m
|
|
23
|
-
Size: [32m[1m14.
|
|
23
|
+
Size: [32m[1m14.49 kB[22m[39m [90mbrotlied[39m
|
|
24
24
|
|
|
25
25
|
[1mdist/index.mjs[22m
|
|
26
26
|
Size limit: [32m[1m50 kB[22m[39m
|
|
27
|
-
Size: [32m[1m14.
|
|
27
|
+
Size: [32m[1m14.37 kB[22m[39m [90mbrotlied[39m
|
|
28
28
|
|
package/dist/index.cjs
CHANGED
|
@@ -1327,6 +1327,96 @@ var models = {
|
|
|
1327
1327
|
tags: ["low-cost", "general-purpose"],
|
|
1328
1328
|
lifecycle: "live"
|
|
1329
1329
|
},
|
|
1330
|
+
"xai:grok-code-fast-1": {
|
|
1331
|
+
id: "xai:grok-code-fast-1",
|
|
1332
|
+
name: "Grok Code Fast 1",
|
|
1333
|
+
description: "Fast coding-optimized Grok model with large context window.",
|
|
1334
|
+
input: {
|
|
1335
|
+
maxTokens: 256e3,
|
|
1336
|
+
costPer1MTokens: 0.2
|
|
1337
|
+
},
|
|
1338
|
+
output: {
|
|
1339
|
+
maxTokens: 32768,
|
|
1340
|
+
costPer1MTokens: 1.5
|
|
1341
|
+
},
|
|
1342
|
+
tags: ["coding", "general-purpose", "low-cost"],
|
|
1343
|
+
lifecycle: "live"
|
|
1344
|
+
},
|
|
1345
|
+
"xai:grok-4-fast-reasoning": {
|
|
1346
|
+
id: "xai:grok-4-fast-reasoning",
|
|
1347
|
+
name: "Grok 4 Fast (Reasoning)",
|
|
1348
|
+
description: "Advanced fast Grok model with reasoning and very large context.",
|
|
1349
|
+
input: {
|
|
1350
|
+
maxTokens: 2e6,
|
|
1351
|
+
costPer1MTokens: 0.2
|
|
1352
|
+
},
|
|
1353
|
+
output: {
|
|
1354
|
+
maxTokens: 128e3,
|
|
1355
|
+
costPer1MTokens: 0.5
|
|
1356
|
+
},
|
|
1357
|
+
tags: ["reasoning", "recommended", "general-purpose"],
|
|
1358
|
+
lifecycle: "live"
|
|
1359
|
+
},
|
|
1360
|
+
"xai:grok-4-fast-non-reasoning": {
|
|
1361
|
+
id: "xai:grok-4-fast-non-reasoning",
|
|
1362
|
+
name: "Grok 4 Fast (Non-Reasoning)",
|
|
1363
|
+
description: "Fast, cost-effective Grok model for non-reasoning tasks.",
|
|
1364
|
+
input: {
|
|
1365
|
+
maxTokens: 2e6,
|
|
1366
|
+
costPer1MTokens: 0.2
|
|
1367
|
+
},
|
|
1368
|
+
output: {
|
|
1369
|
+
maxTokens: 128e3,
|
|
1370
|
+
costPer1MTokens: 0.5
|
|
1371
|
+
},
|
|
1372
|
+
tags: ["low-cost", "recommended", "general-purpose"],
|
|
1373
|
+
lifecycle: "live"
|
|
1374
|
+
},
|
|
1375
|
+
"xai:grok-4-0709": {
|
|
1376
|
+
id: "xai:grok-4-0709",
|
|
1377
|
+
name: "Grok 4 (0709)",
|
|
1378
|
+
description: "Comprehensive Grok 4 model for general-purpose tasks.",
|
|
1379
|
+
input: {
|
|
1380
|
+
maxTokens: 256e3,
|
|
1381
|
+
costPer1MTokens: 3
|
|
1382
|
+
},
|
|
1383
|
+
output: {
|
|
1384
|
+
maxTokens: 32768,
|
|
1385
|
+
costPer1MTokens: 15
|
|
1386
|
+
},
|
|
1387
|
+
tags: ["reasoning", "general-purpose"],
|
|
1388
|
+
lifecycle: "live"
|
|
1389
|
+
},
|
|
1390
|
+
"xai:grok-3-mini": {
|
|
1391
|
+
id: "xai:grok-3-mini",
|
|
1392
|
+
name: "Grok 3 Mini",
|
|
1393
|
+
description: "Lightweight Grok model for cost-sensitive workloads.",
|
|
1394
|
+
input: {
|
|
1395
|
+
maxTokens: 131072,
|
|
1396
|
+
costPer1MTokens: 0.3
|
|
1397
|
+
},
|
|
1398
|
+
output: {
|
|
1399
|
+
maxTokens: 16384,
|
|
1400
|
+
costPer1MTokens: 0.5
|
|
1401
|
+
},
|
|
1402
|
+
tags: ["low-cost", "general-purpose"],
|
|
1403
|
+
lifecycle: "live"
|
|
1404
|
+
},
|
|
1405
|
+
"xai:grok-3": {
|
|
1406
|
+
id: "xai:grok-3",
|
|
1407
|
+
name: "Grok 3",
|
|
1408
|
+
description: "Enterprise-grade Grok model for general-purpose tasks.",
|
|
1409
|
+
input: {
|
|
1410
|
+
maxTokens: 131072,
|
|
1411
|
+
costPer1MTokens: 3
|
|
1412
|
+
},
|
|
1413
|
+
output: {
|
|
1414
|
+
maxTokens: 16384,
|
|
1415
|
+
costPer1MTokens: 15
|
|
1416
|
+
},
|
|
1417
|
+
tags: ["general-purpose"],
|
|
1418
|
+
lifecycle: "live"
|
|
1419
|
+
},
|
|
1330
1420
|
"openrouter:gpt-oss-120b": {
|
|
1331
1421
|
id: "openrouter:gpt-oss-120b",
|
|
1332
1422
|
name: "GPT-OSS 120B (Preview)",
|
|
@@ -1342,8 +1432,8 @@ var models = {
|
|
|
1342
1432
|
tags: ["preview", "general-purpose", "reasoning"],
|
|
1343
1433
|
lifecycle: "live"
|
|
1344
1434
|
},
|
|
1345
|
-
"fireworks:gpt-oss-20b": {
|
|
1346
|
-
id: "fireworks:gpt-oss-20b",
|
|
1435
|
+
"fireworks-ai:gpt-oss-20b": {
|
|
1436
|
+
id: "fireworks-ai:gpt-oss-20b",
|
|
1347
1437
|
name: "GPT-OSS 20B",
|
|
1348
1438
|
description: "gpt-oss-20b is a compact, open-weight language model optimized for low-latency. It shares the same training foundation and capabilities as the GPT-OSS 120B model, with faster responses and lower cost.",
|
|
1349
1439
|
input: {
|
|
@@ -1358,8 +1448,8 @@ var models = {
|
|
|
1358
1448
|
lifecycle: "live",
|
|
1359
1449
|
aliases: ["accounts/fireworks/models/gpt-oss-20b"]
|
|
1360
1450
|
},
|
|
1361
|
-
"fireworks:gpt-oss-120b": {
|
|
1362
|
-
id: "fireworks:gpt-oss-120b",
|
|
1451
|
+
"fireworks-ai:gpt-oss-120b": {
|
|
1452
|
+
id: "fireworks-ai:gpt-oss-120b",
|
|
1363
1453
|
name: "GPT-OSS 120B",
|
|
1364
1454
|
description: "gpt-oss-120b is a high-performance, open-weight language model designed for production-grade, general-purpose use cases. It excels at complex reasoning and supports configurable reasoning effort, full chain-of-thought transparency for easier debugging and trust, and native agentic capabilities for function calling, tool use, and structured outputs.",
|
|
1365
1455
|
input: {
|
|
@@ -1374,8 +1464,8 @@ var models = {
|
|
|
1374
1464
|
lifecycle: "live",
|
|
1375
1465
|
aliases: ["accounts/fireworks/models/gpt-oss-120b"]
|
|
1376
1466
|
},
|
|
1377
|
-
"fireworks:deepseek-r1-0528": {
|
|
1378
|
-
id: "fireworks:deepseek-r1-0528",
|
|
1467
|
+
"fireworks-ai:deepseek-r1-0528": {
|
|
1468
|
+
id: "fireworks-ai:deepseek-r1-0528",
|
|
1379
1469
|
name: "DeepSeek R1 0528",
|
|
1380
1470
|
description: "The updated DeepSeek R1 0528 model delivers major improvements in reasoning, inference, and accuracy through enhanced post-training optimization and greater computational resources. It now performs at a level approaching top-tier models like OpenAI o3 and Gemini 2.5 Pro, with notable gains in complex tasks such as math and programming. The update also reduces hallucinations, improves function calling, and enhances the coding experience.",
|
|
1381
1471
|
input: {
|
|
@@ -1390,8 +1480,8 @@ var models = {
|
|
|
1390
1480
|
lifecycle: "live",
|
|
1391
1481
|
aliases: ["accounts/fireworks/models/deepseek-r1-0528"]
|
|
1392
1482
|
},
|
|
1393
|
-
"fireworks:deepseek-v3-0324": {
|
|
1394
|
-
id: "fireworks:deepseek-v3-0324",
|
|
1483
|
+
"fireworks-ai:deepseek-v3-0324": {
|
|
1484
|
+
id: "fireworks-ai:deepseek-v3-0324",
|
|
1395
1485
|
name: "DeepSeek V3 0324",
|
|
1396
1486
|
description: "DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the DeepSeek V3 model and performs really well on a variety of tasks.",
|
|
1397
1487
|
input: {
|
|
@@ -1406,8 +1496,8 @@ var models = {
|
|
|
1406
1496
|
lifecycle: "live",
|
|
1407
1497
|
aliases: ["accounts/fireworks/models/deepseek-v3-0324"]
|
|
1408
1498
|
},
|
|
1409
|
-
"fireworks:llama4-maverick-instruct-basic": {
|
|
1410
|
-
id: "fireworks:llama4-maverick-instruct-basic",
|
|
1499
|
+
"fireworks-ai:llama4-maverick-instruct-basic": {
|
|
1500
|
+
id: "fireworks-ai:llama4-maverick-instruct-basic",
|
|
1411
1501
|
name: "Llama 4 Maverick Instruct (Basic)",
|
|
1412
1502
|
description: "Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward pass (400B total). It supports multilingual text and image input, and produces multilingual text and code output across 12 supported languages. Optimized for vision-language tasks, Maverick is instruction-tuned for assistant-like behavior, image reasoning, and general-purpose multimodal interaction, and suited for research and commercial applications requiring advanced multimodal understanding and high model throughput.",
|
|
1413
1503
|
input: {
|
|
@@ -1422,8 +1512,8 @@ var models = {
|
|
|
1422
1512
|
lifecycle: "live",
|
|
1423
1513
|
aliases: ["accounts/fireworks/models/llama4-maverick-instruct-basic"]
|
|
1424
1514
|
},
|
|
1425
|
-
"fireworks:llama4-scout-instruct-basic": {
|
|
1426
|
-
id: "fireworks:llama4-scout-instruct-basic",
|
|
1515
|
+
"fireworks-ai:llama4-scout-instruct-basic": {
|
|
1516
|
+
id: "fireworks-ai:llama4-scout-instruct-basic",
|
|
1427
1517
|
name: "Llama 4 Scout Instruct (Basic)",
|
|
1428
1518
|
description: "Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, uses 16 experts per forward pass, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, it is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks.",
|
|
1429
1519
|
input: {
|
|
@@ -1438,8 +1528,8 @@ var models = {
|
|
|
1438
1528
|
lifecycle: "live",
|
|
1439
1529
|
aliases: ["accounts/fireworks/models/llama4-scout-instruct-basic"]
|
|
1440
1530
|
},
|
|
1441
|
-
"fireworks:llama-v3p3-70b-instruct": {
|
|
1442
|
-
id: "fireworks:llama-v3p3-70b-instruct",
|
|
1531
|
+
"fireworks-ai:llama-v3p3-70b-instruct": {
|
|
1532
|
+
id: "fireworks-ai:llama-v3p3-70b-instruct",
|
|
1443
1533
|
name: "Llama 3.3 70B Instruct",
|
|
1444
1534
|
description: "Llama 3.3 70B Instruct is the December update of Llama 3.1 70B. The model improves upon Llama 3.1 70B (released July 2024) with advances in tool calling, multilingual text support, math and coding. The model achieves industry leading results in reasoning, math and instruction following and provides similar performance as 3.1 405B but with significant speed and cost improvements.",
|
|
1445
1535
|
input: {
|
|
@@ -1454,8 +1544,8 @@ var models = {
|
|
|
1454
1544
|
lifecycle: "live",
|
|
1455
1545
|
aliases: ["accounts/fireworks/models/llama-v3p3-70b-instruct"]
|
|
1456
1546
|
},
|
|
1457
|
-
"fireworks:deepseek-r1": {
|
|
1458
|
-
id: "fireworks:deepseek-r1",
|
|
1547
|
+
"fireworks-ai:deepseek-r1": {
|
|
1548
|
+
id: "fireworks-ai:deepseek-r1",
|
|
1459
1549
|
name: "DeepSeek R1 (Fast)",
|
|
1460
1550
|
description: "This version of the R1 model has a perfect balance between speed and cost-efficiency for real-time interactive experiences, with speeds up to 90 tokens per second.\n\nDeepSeek-R1 is a state-of-the-art large language model optimized with reinforcement learning and cold-start data for exceptional reasoning, math, and code performance. **Note**: This model will always use a temperature of 0.6 as recommended by DeepSeek.",
|
|
1461
1551
|
input: {
|
|
@@ -1470,8 +1560,8 @@ var models = {
|
|
|
1470
1560
|
lifecycle: "live",
|
|
1471
1561
|
aliases: ["accounts/fireworks/models/deepseek-r1"]
|
|
1472
1562
|
},
|
|
1473
|
-
"fireworks:deepseek-r1-basic": {
|
|
1474
|
-
id: "fireworks:deepseek-r1-basic",
|
|
1563
|
+
"fireworks-ai:deepseek-r1-basic": {
|
|
1564
|
+
id: "fireworks-ai:deepseek-r1-basic",
|
|
1475
1565
|
name: "DeepSeek R1 (Basic)",
|
|
1476
1566
|
description: 'This version of the R1 model is optimized for throughput and cost-effectiveness and has a lower cost but slightly higher latency than the "Fast" version of the model.\n\nDeepSeek-R1 is a state-of-the-art large language model optimized with reinforcement learning and cold-start data for exceptional reasoning, math, and code performance. **Note**: This model will always use a temperature of 0.6 as recommended by DeepSeek.',
|
|
1477
1567
|
input: {
|
|
@@ -1486,8 +1576,8 @@ var models = {
|
|
|
1486
1576
|
lifecycle: "live",
|
|
1487
1577
|
aliases: ["accounts/fireworks/models/deepseek-r1-basic"]
|
|
1488
1578
|
},
|
|
1489
|
-
"fireworks:deepseek-v3": {
|
|
1490
|
-
id: "fireworks:deepseek-v3",
|
|
1579
|
+
"fireworks-ai:deepseek-v3": {
|
|
1580
|
+
id: "fireworks-ai:deepseek-v3",
|
|
1491
1581
|
name: "DeepSeek V3",
|
|
1492
1582
|
description: "A a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token from Deepseek.",
|
|
1493
1583
|
input: {
|
|
@@ -1502,8 +1592,8 @@ var models = {
|
|
|
1502
1592
|
lifecycle: "deprecated",
|
|
1503
1593
|
aliases: ["accounts/fireworks/models/deepseek-v3"]
|
|
1504
1594
|
},
|
|
1505
|
-
"fireworks:llama-v3p1-405b-instruct": {
|
|
1506
|
-
id: "fireworks:llama-v3p1-405b-instruct",
|
|
1595
|
+
"fireworks-ai:llama-v3p1-405b-instruct": {
|
|
1596
|
+
id: "fireworks-ai:llama-v3p1-405b-instruct",
|
|
1507
1597
|
name: "Llama 3.1 405B Instruct",
|
|
1508
1598
|
description: "The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes. The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.",
|
|
1509
1599
|
input: {
|
|
@@ -1518,8 +1608,8 @@ var models = {
|
|
|
1518
1608
|
lifecycle: "deprecated",
|
|
1519
1609
|
aliases: ["accounts/fireworks/models/llama-v3p1-405b-instruct"]
|
|
1520
1610
|
},
|
|
1521
|
-
"fireworks:llama-v3p1-70b-instruct": {
|
|
1522
|
-
id: "fireworks:llama-v3p1-70b-instruct",
|
|
1611
|
+
"fireworks-ai:llama-v3p1-70b-instruct": {
|
|
1612
|
+
id: "fireworks-ai:llama-v3p1-70b-instruct",
|
|
1523
1613
|
name: "Llama 3.1 70B Instruct",
|
|
1524
1614
|
description: "The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes. The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.",
|
|
1525
1615
|
input: {
|
|
@@ -1534,8 +1624,8 @@ var models = {
|
|
|
1534
1624
|
lifecycle: "deprecated",
|
|
1535
1625
|
aliases: ["accounts/fireworks/models/llama-v3p1-70b-instruct"]
|
|
1536
1626
|
},
|
|
1537
|
-
"fireworks:llama-v3p1-8b-instruct": {
|
|
1538
|
-
id: "fireworks:llama-v3p1-8b-instruct",
|
|
1627
|
+
"fireworks-ai:llama-v3p1-8b-instruct": {
|
|
1628
|
+
id: "fireworks-ai:llama-v3p1-8b-instruct",
|
|
1539
1629
|
name: "Llama 3.1 8B Instruct",
|
|
1540
1630
|
description: "The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes. The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.",
|
|
1541
1631
|
input: {
|
|
@@ -1550,8 +1640,8 @@ var models = {
|
|
|
1550
1640
|
lifecycle: "live",
|
|
1551
1641
|
aliases: ["accounts/fireworks/models/llama-v3p1-8b-instruct"]
|
|
1552
1642
|
},
|
|
1553
|
-
"fireworks:mixtral-8x22b-instruct": {
|
|
1554
|
-
id: "fireworks:mixtral-8x22b-instruct",
|
|
1643
|
+
"fireworks-ai:mixtral-8x22b-instruct": {
|
|
1644
|
+
id: "fireworks-ai:mixtral-8x22b-instruct",
|
|
1555
1645
|
name: "Mixtral MoE 8x22B Instruct",
|
|
1556
1646
|
description: "Mistral MoE 8x22B Instruct v0.1 model with Sparse Mixture of Experts. Fine tuned for instruction following.",
|
|
1557
1647
|
input: {
|
|
@@ -1566,8 +1656,8 @@ var models = {
|
|
|
1566
1656
|
lifecycle: "live",
|
|
1567
1657
|
aliases: ["accounts/fireworks/models/mixtral-8x22b-instruct"]
|
|
1568
1658
|
},
|
|
1569
|
-
"fireworks:mixtral-8x7b-instruct": {
|
|
1570
|
-
id: "fireworks:mixtral-8x7b-instruct",
|
|
1659
|
+
"fireworks-ai:mixtral-8x7b-instruct": {
|
|
1660
|
+
id: "fireworks-ai:mixtral-8x7b-instruct",
|
|
1571
1661
|
name: "Mixtral MoE 8x7B Instruct",
|
|
1572
1662
|
description: "Mistral MoE 8x7B Instruct v0.1 model with Sparse Mixture of Experts. Fine tuned for instruction following",
|
|
1573
1663
|
input: {
|
|
@@ -1582,8 +1672,8 @@ var models = {
|
|
|
1582
1672
|
lifecycle: "live",
|
|
1583
1673
|
aliases: ["accounts/fireworks/models/mixtral-8x7b-instruct"]
|
|
1584
1674
|
},
|
|
1585
|
-
"fireworks:mythomax-l2-13b": {
|
|
1586
|
-
id: "fireworks:mythomax-l2-13b",
|
|
1675
|
+
"fireworks-ai:mythomax-l2-13b": {
|
|
1676
|
+
id: "fireworks-ai:mythomax-l2-13b",
|
|
1587
1677
|
name: "MythoMax L2 13b",
|
|
1588
1678
|
description: "MythoMax L2 is designed to excel at both roleplaying and storytelling, and is an improved variant of the previous MythoMix model, combining the MythoLogic-L2 and Huginn models.",
|
|
1589
1679
|
input: {
|
|
@@ -1598,8 +1688,8 @@ var models = {
|
|
|
1598
1688
|
lifecycle: "live",
|
|
1599
1689
|
aliases: ["accounts/fireworks/models/mythomax-l2-13b"]
|
|
1600
1690
|
},
|
|
1601
|
-
"fireworks:gemma2-9b-it": {
|
|
1602
|
-
id: "fireworks:gemma2-9b-it",
|
|
1691
|
+
"fireworks-ai:gemma2-9b-it": {
|
|
1692
|
+
id: "fireworks-ai:gemma2-9b-it",
|
|
1603
1693
|
name: "Gemma 2 9B Instruct",
|
|
1604
1694
|
description: "Redesigned for outsized performance and unmatched efficiency, Gemma 2 optimizes for blazing-fast inference on diverse hardware. Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. They are text-to-text, decoder-only large language models, available in English, with open weights, pre-trained variants, and instruction-tuned variants. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.",
|
|
1605
1695
|
input: {
|